<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	 xmlns:media="http://search.yahoo.com/mrss/" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:media="http://search.yahoo.com/mrss/"
>

<channel>
	<title>RPABOTS.WORLD</title>
	<atom:link href="https://rpabotsworld.com/tag/crewai/feed/" rel="self" type="application/rss+xml" />
	<link>https://rpabotsworld.com</link>
	<description>RPA, Agentic AI &amp; Intelligent Automation — Tutorials, Tools &amp; Career Guides</description>
	<lastBuildDate>Fri, 30 Oct 2026 02:55:35 +0000</lastBuildDate>
	<language>en-US</language>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=7.1.5</generator>
	<itunes:subtitle>RPABOTS.WORLD</itunes:subtitle>
	<itunes:summary>RPA, Agentic AI &amp; Intelligent Automation — Tutorials, Tools &amp; Career Guides</itunes:summary>
	<itunes:explicit>clean</itunes:explicit>
	<item>
		<title>LangGraph vs. CrewAI vs. Microsoft Agent Framework vs. Google ADK: Which Multi-Agent Framework Should You Actually Learn in 2026?</title>
		<link>https://rpabotsworld.com/langgraph-vs-crewai-vs-microsoft-agent-framework-vs-google-adk/</link>
					<comments>https://rpabotsworld.com/langgraph-vs-crewai-vs-microsoft-agent-framework-vs-google-adk/#comments</comments>
		
		<dc:creator><![CDATA[Satish Prasad]]></dc:creator>
		<pubDate>Tue, 14 Jul 2026 02:47:57 +0000</pubDate>
				<category><![CDATA[Multi-Agent Systems]]></category>
		<category><![CDATA[best AI agent framework 2026]]></category>
		<category><![CDATA[crewai]]></category>
		<category><![CDATA[Google ADK multi-agent]]></category>
		<category><![CDATA[LangGraph]]></category>
		<category><![CDATA[LangGraph vs CrewAI]]></category>
		<category><![CDATA[Microsoft Agent Framework vs AutoGen]]></category>
		<guid isPermaLink="false">https://rpabotsworld.com/?p=32166</guid>

					<description><![CDATA[Ask five AI engineers which multi-agent framework to learn and you&#8217;ll get five different answers, three of which are now technically wrong. AutoGen — the framework half the internet&#8217;s tutorials are still written against — went into maintenance mode in 2026. Microsoft folded it into a new SDK. Semantic Kernel, its enterprise sibling, went with [&#8230;]]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Ask five AI engineers which multi-agent framework to learn and you&#8217;ll get five different answers, three of which are now technically wrong. AutoGen — the framework half the internet&#8217;s tutorials are still written against — went into maintenance mode in 2026. Microsoft folded it into a new SDK. Semantic Kernel, its enterprise sibling, went with it. If you bookmarked a &#8220;getting started with AutoGen&#8221; guide sometime in the last two years, it&#8217;s describing a framework Microsoft itself no longer wants you to start new projects on.</p>



<p class="wp-block-paragraph">That&#8217;s the problem with this space: the frameworks move faster than the content written about them. So instead of another feature-matrix post, this is a decision guide — what each of these four actually is, how they differ architecturally (not just in syntax), and a direct verdict on which one fits your situation. If you want the deeper technical background on any one of them, we&#8217;ve covered each in dedicated guides linked throughout — this is the piece that ties them together.</p>



<h2 class="wp-block-heading">The Decision Table</h2>



<figure class="wp-block-table"><table class="has-fixed-layout"><thead><tr><th></th><th><strong>LangGraph</strong></th><th><strong>CrewAI</strong></th><th><strong>Microsoft Agent Framework</strong></th><th><strong>Google ADK</strong></th></tr></thead><tbody><tr><td><strong>Core mental model</strong></td><td>Agents as nodes in a state graph you control explicitly</td><td>Agents as role-based team members (&#8220;crew&#8221;) delegating tasks</td><td>Agents composed via orchestration patterns: sequential, concurrent, handoff, group chat, Magentic-One</td><td>Agents composed via workflow types: template (Sequential/Parallel/Loop), graph-based, or dynamic coordinator + sub-agents</td></tr><tr><td><strong>Best for</strong></td><td>Production-grade, long-running, stateful agents where you need precise control over execution flow</td><td>Fast-to-build teams of specialized agents with clear task delegation</td><td>Enterprise .NET/Python shops already on Semantic Kernel or migrating off AutoGen</td><td>Teams building on Google Cloud / Vertex AI, or wanting one framework across Python, TypeScript, Go, Java, and Kotlin</td></tr><tr><td><strong>Learning curve</strong></td><td>Steepest — it&#8217;s a low-level orchestration runtime, not a batteries-included agent framework</td><td>Lowest — role-based DSL, a working crew in ~20 lines</td><td>Moderate — familiar if you know Semantic Kernel; new if you only knew AutoGen</td><td>Moderate — template workflow agents are quick to start, graph workflows add depth</td></tr><tr><td><strong>State &amp; durability</strong></td><td>Built-in checkpointing, durable execution, human-in-the-loop as a first-class concept</td><td>Lighter-weight; state passes through crew/task context</td><td>Type-safe workflow engine with checkpointing, carried over from Semantic Kernel&#8217;s enterprise plumbing</td><td>Sessions, State, Memory, and Events as distinct built-in primitives</td></tr><tr><td><strong>Protocol support</strong></td><td>MCP-compatible via LangChain tool integrations</td><td>Growing MCP support</td><td>Native MCP <strong>and</strong> A2A (Agent-to-Agent) protocol support</td><td>Native MCP tools and A2A protocol support (both directions: expose and consume)</td></tr><tr><td><strong>Deployment target</strong></td><td>Self-hosted, LangGraph Platform, or any cloud runtime</td><td>Self-hosted or CrewAI&#8217;s own platform</td><td>Azure-native, but framework itself is open-source and portable</td><td>Vertex AI Agent Engine, Cloud Run, GKE, or self-hosted</td></tr><tr><td><strong>Maturity signal (mid-2026)</strong></td><td>Actively developed; recommended baseline is 0.4+ for checkpointing and observability</td><td>Actively developed; recommended baseline is 0.105+</td><td>Shipped 1.0 production release in April 2026 — genuinely new, not battle-tested at scale yet</td><td>ADK 2.0 shipped with graph workflows in 2026; multi-language SDKs are maturing at different rates</td></tr><tr><td><strong>Who&#8217;s behind it</strong></td><td>LangChain</td><td>CrewAI Inc.</td><td>Microsoft (unifying the AutoGen and Semantic Kernel teams)</td><td>Google</td></tr></tbody></table></figure>



<p class="wp-block-paragraph">If you only read one row, read &#8220;core mental model&#8221; — it predicts almost everything else about how a framework will feel to build in.</p>



<h2 class="wp-block-heading">LangGraph: The Low-Level Orchestration Runtime</h2>



<p class="wp-block-paragraph">LangGraph is not, strictly speaking, an &#8220;agent framework&#8221; in the same sense as the other three — it&#8217;s a low-level orchestration runtime that agent frameworks (including LangChain&#8217;s own <code>create_agent</code>) get built on top of. LangChain&#8217;s own documentation is explicit about this: &#8220;LangGraph is very low-level, and focused entirely on agent orchestration,&#8221; recommending that if you want a higher-level abstraction, you start with LangChain&#8217;s prebuilt agent loops instead and drop into LangGraph when you need explicit control. It&#8217;s trusted in production by Klarna, Uber, and J.P. Morgan — the kind of adopters who needed the durable execution and checkpointing that a request-response chatbot doesn&#8217;t. (<a href="https://docs.langchain.com/oss/python/langgraph/overview" rel="nofollow noopener" target="_blank">LangChain, LangGraph Overview</a>)</p>



<p class="wp-block-paragraph">For multi-agent systems specifically, LangGraph (via LangChain&#8217;s agent abstractions) supports several distinct patterns rather than one fixed structure:</p>



<ul class="wp-block-list">
<li><strong>Subagents</strong> — a main agent coordinates specialist subagents as tools; all routing passes through the main agent.</li>



<li><strong>Handoffs</strong> — agents dynamically hand off control to each other via state-changing tool calls (the pattern our <a href="https://rpabotsworld.com/build-deploy-langgraph-agent-uipath/">HTS classification agent</a> tutorial uses for escalation).</li>



<li><strong>Skills</strong> — a single agent stays in control while loading specialized prompts/knowledge on demand.</li>



<li><strong>Router</strong> — a classification step directs input to one or more specialist agents, then synthesizes the results.</li>



<li><strong>Custom workflow</strong> — you build the graph by hand, mixing deterministic logic with agentic steps.</li>
</ul>



<p class="wp-block-paragraph">Notably, LangChain has been migrating away from the older <code>langgraph-supervisor</code> package (which modeled worker agents as separate graph nodes routed by a supervisor) toward wrapping each worker as a callable tool on a coordinating agent — a subtle but real architectural shift worth knowing if you&#8217;re following an older tutorial. (<a href="https://docs.langchain.com/oss/python/migrate/langgraph-supervisor" rel="nofollow noopener" target="_blank">LangChain, Supervisor Migration Guide</a>)</p>



<p class="wp-block-paragraph">The trade-off is honest: nothing here is &#8220;20 lines to a working crew.&#8221; You&#8217;re building a state machine, and you own every edge. That&#8217;s exactly what makes it the right choice when correctness, auditability, and long-running state matter more than speed of first prototype — which is why it&#8217;s the framework we reach for in UiPath deployment tutorials where an audit trail is non-negotiable.</p>



<h2 class="wp-block-heading">CrewAI: Role-Based Teams, Fast to Build</h2>



<p class="wp-block-paragraph">CrewAI&#8217;s mental model is the most intuitive of the four: you define agents with roles (&#8220;Research Analyst,&#8221; &#8220;Content Writer&#8221;), give them goals and backstories, assign them tasks, and CrewAI&#8217;s crew orchestration handles delegation between them. It consistently comes up as the framework with the lowest learning curve — a working multi-agent crew in roughly 20 lines is a realistic claim, not a marketing one. (<a href="https://www.datacamp.com/tutorial/crewai-vs-langgraph-vs-autogen" rel="nofollow noopener" target="_blank">DataCamp, CrewAI vs LangGraph vs AutoGen</a>)</p>



<p class="wp-block-paragraph">We&#8217;ve covered CrewAI&#8217;s memory architecture and Knowledge feature in depth elsewhere — see <a href="https://rpabotsworld.com/crewai-memory-types/">Unlocking CrewAI Memory Types</a> and <a href="https://rpabotsworld.com/crewai-knowledge-feature/">Unlocking CrewAI Knowledge Feature</a> — and both remain accurate descriptions of an actively developed framework. That matters more than it sounds: unlike AutoGen, CrewAI hasn&#8217;t been folded into anything else, and community momentum has genuinely shifted toward it as an AutoGen alternative since Microsoft&#8217;s announcement.</p>



<p class="wp-block-paragraph">CrewAI&#8217;s ceiling is lower than LangGraph&#8217;s for genuinely complex, branching, long-running workflows — the role/task/crew abstraction starts to feel restrictive once you need fine-grained control over exactly when and how agents communicate. But for the common case — &#8220;I need three or four specialized agents to divide up a task and hand back a result&#8221; — it gets you there fastest.</p>



<h2 class="wp-block-heading">Microsoft Agent Framework: AutoGen and Semantic Kernel, Merged</h2>



<p class="wp-block-paragraph">This is the framework most likely to surprise readers who last looked at this space a year ago. On April 3, 2026, Microsoft shipped <strong>Microsoft Agent Framework 1.0</strong> — a production-ready unification of AutoGen (Microsoft Research&#8217;s multi-agent conversation framework) and Semantic Kernel (Microsoft&#8217;s enterprise-grade agent SDK) into a single framework, following a public preview in October 2025. Microsoft&#8217;s own documentation calls it &#8220;the direct successor&#8221; to both, built by the same teams. (<a href="https://devblogs.microsoft.com/agent-framework/microsoft-agent-framework-version-1-0/" rel="nofollow noopener" target="_blank">Microsoft Agent Framework 1.0 announcement</a>, <a href="https://visualstudiomagazine.com/articles/2026/04/06/microsoft-ships-production-ready-agent-framework-1-0-for-net-and-python.aspx" rel="nofollow noopener" target="_blank">Visual Studio Magazine</a>)</p>



<p class="wp-block-paragraph">Architecturally, it takes Semantic Kernel&#8217;s enterprise plumbing — session-based state management, type safety, filters, telemetry, and its connector ecosystem — and combines it with the multi-agent orchestration patterns that made AutoGen interesting: sequential, concurrent, handoff, group chat, and the more sophisticated Magentic-One pattern for open-ended, dynamically planned tasks. On top of that sits a graph-based workflow engine with type-safe routing and checkpointing. It supports both MCP (for tool discovery) and Google&#8217;s A2A protocol (for cross-runtime agent-to-agent collaboration) natively. (<a href="https://devblogs.microsoft.com/agent-framework/migrate-your-semantic-kernel-and-autogen-projects-to-microsoft-agent-framework-release-candidate/" rel="nofollow noopener" target="_blank">Microsoft Agent Framework migration guide</a>)</p>



<p class="wp-block-paragraph">Two practical implications:</p>



<ol class="wp-block-list">
<li><strong>If you&#8217;re on AutoGen today, you have a migration to plan.</strong> It&#8217;s not abandoned — bug fixes and security patches continue — but new feature investment is going into Agent Framework, and Microsoft is actively publishing migration guides.</li>



<li><strong>If you&#8217;re a Microsoft-stack shop</strong>, this is now the natural on-ramp — it sits next to <a href="https://rpabotsworld.com/microsoft-copilot-the-complete-guide/">Microsoft Copilot</a> and Azure AI Foundry in the same ecosystem, with .NET and Python parity from day one.</li>
</ol>



<p class="wp-block-paragraph">It&#8217;s genuinely new as a unified 1.0 product, though — &#8220;production-ready&#8221; per Microsoft&#8217;s release notes doesn&#8217;t mean &#8220;battle-tested at the scale LangGraph or CrewAI have accumulated.&#8221; Weigh that if you&#8217;re choosing for a large existing team versus a greenfield Microsoft-native project.</p>



<h2 class="wp-block-heading">Google ADK: Multi-Language, Multi-Pattern, Cloud-Native</h2>



<p class="wp-block-paragraph">Google&#8217;s Agent Development Kit stands out for two reasons: it&#8217;s genuinely multi-language (Python, TypeScript, Go, Java, and Kotlin SDKs, with Go reaching 2.0 GA in 2026), and it offers the widest range of explicit workflow structures of any framework here. ADK gives you four distinct ways to compose multi-agent systems: <strong>template workflows</strong> (<code>SequentialAgent</code>, <code>ParallelAgent</code>, and <code>LoopAgent</code> — pre-built patterns extending a common <code>BaseAgent</code>), <strong>graph-based workflows</strong> (ADK 2.0+, mixing AI agents and deterministic nodes with branching), <strong>dynamic workflows</strong> (full programmatic control), and <strong>collaborative workflows</strong> (a single coordinator agent dynamically directing a set of sub-agents). (<a href="https://adk.dev/workflows/" rel="nofollow noopener" target="_blank">Google ADK, Workflows</a>)</p>



<p class="wp-block-paragraph">A common ADK pattern looks structurally similar to AutoGen&#8217;s GroupChat or a CrewAI crew — a planner dispatches to a retrieval sub-agent, then a drafting sub-agent, then a grading sub-agent — but ADK&#8217;s distinguishing trait is that workflow agents are themselves agents, so they nest: a <code>SequentialAgent</code> can contain a <code>ParallelAgent</code>, which can contain another <code>SequentialAgent</code>. (<a href="https://developers.googleblog.com/developers-guide-to-multi-agent-patterns-in-adk/" rel="nofollow noopener" target="_blank">Google Developers Blog, Multi-Agent Patterns in ADK</a>)</p>



<p class="wp-block-paragraph">We covered building a production multi-agent system with ADK in detail in <a href="https://rpabotsworld.com/building-multi-agent-systems-with-google-adk-the-complete-step-by-step-guide/">Building Multi-Agent Systems with Google ADK</a> — that guide holds up well against ADK&#8217;s current architecture. Where ADK pulls ahead of the other three is deployment flexibility if you&#8217;re already committed to Google Cloud: first-class paths to Vertex AI Agent Engine, Cloud Run, and GKE, plus a built-in evaluation framework and both MCP and A2A protocol support in both directions (exposing your agents to other systems and consuming external agents).</p>



<h2 class="wp-block-heading">Orchestration Patterns, Side by Side</h2>



<p class="wp-block-paragraph">The four frameworks converge on a similar set of underlying patterns even though the terminology differs — worth knowing so you can translate a tutorial written for one framework into another:</p>



<figure class="wp-block-table"><table class="has-fixed-layout"><thead><tr><th>Pattern</th><th>LangGraph</th><th>CrewAI</th><th>Microsoft Agent Framework</th><th>Google ADK</th></tr></thead><tbody><tr><td>Fixed sequence of agents</td><td>Custom workflow / sequential edges</td><td>Sequential process</td><td>Sequential orchestration</td><td><code>SequentialAgent</code></td></tr><tr><td>Agents run concurrently</td><td>Custom workflow (parallel nodes)</td><td>Not a first-class primitive</td><td>Concurrent orchestration</td><td><code>ParallelAgent</code></td></tr><tr><td>Agent-to-agent dynamic handoff</td><td>Handoffs pattern</td><td>Delegation via manager agent</td><td>Handoff orchestration</td><td>Agent routing (experimental)</td></tr><tr><td>Open-ended group discussion</td><td>Custom graph</td><td>Not a native pattern</td><td>Group chat / Magentic-One</td><td>Collaborative workflows</td></tr><tr><td>Repeat-until-condition</td><td>Custom workflow with a loop edge</td><td>Not a native primitive</td><td>Workflow engine loop</td><td><code>LoopAgent</code></td></tr></tbody></table></figure>



<h2 class="wp-block-heading">Production Readiness: What to Actually Pin</h2>



<p class="wp-block-paragraph">If you&#8217;re starting a new project in mid-to-late 2026, the safe version floors are LangGraph 0.4 or later and CrewAI 0.105 or later — anything older is missing checkpointing, observability, or current API surfaces you&#8217;ll want within six months. (<a href="https://tensoria.fr/en/blog/multi-agent-orchestration-comparison" rel="nofollow noopener" target="_blank">Tensoria, Multi-Agent Orchestration Comparison</a>) For Microsoft Agent Framework, start at 1.0 — there&#8217;s no meaningful pre-1.0 version worth building on now that AutoGen and Semantic Kernel are both deprecated in favor of it. For Google ADK, use 2.0+ if you want graph-based workflows; the template workflow agents (Sequential/Parallel/Loop) have been stable for longer and are the lower-risk starting point if you don&#8217;t need graph branching yet.</p>



<h2 class="wp-block-heading">The Verdict</h2>



<p class="wp-block-paragraph"><strong>Use LangGraph when</strong> you&#8217;re building a production agent that needs to run for minutes, hours, or days; needs durable state and human-in-the-loop review; and where you&#8217;re willing to trade a steeper learning curve for explicit control over every transition. This is our default recommendation for anything deploying into a governed environment like UiPath Orchestrator, where auditability matters as much as capability — see our <a href="https://rpabotsworld.com/build-deploy-langgraph-agent-uipath/">LangGraph + UiPath deployment tutorial</a> for exactly that pattern.</p>



<p class="wp-block-paragraph"><strong>Use CrewAI when</strong> you want the fastest path from idea to a working multi-agent prototype, your task decomposes naturally into clear roles, and you don&#8217;t yet need fine-grained control over exactly how and when agents communicate. It&#8217;s also a reasonable default if your team is newer to agent development and needs to build intuition before graduating to something lower-level.</p>



<p class="wp-block-paragraph"><strong>Use Microsoft Agent Framework when</strong> you&#8217;re already invested in Semantic Kernel, migrating off AutoGen, or building inside a Microsoft/Azure-native stack where Copilot Studio, Azure AI Foundry, and .NET tooling are already part of your architecture. Go in with eyes open that the unified 1.0 is new — validate it against your specific reliability requirements before betting a large team on it.</p>



<p class="wp-block-paragraph"><strong>Use Google ADK when</strong> you need multi-language support across a polyglot team (Python, TypeScript, Go, Java, Kotlin), you&#8217;re deploying to Vertex AI or Google Cloud generally, or you want the widest built-in menu of workflow structures without hand-rolling a graph. It&#8217;s also the strongest choice if A2A cross-agent interoperability with non-Google systems is a hard requirement.</p>



<p class="wp-block-paragraph">None of these are permanent commitments — MCP and A2A protocol support across all four means agents built on different frameworks can increasingly talk to each other rather than forcing an all-or-nothing bet. Pick based on where your team and infrastructure already are, not on which framework has the loudest hype cycle this quarter.</p>



<h2 class="wp-block-heading">Key Takeaways</h2>



<ul class="wp-block-list">
<li>AutoGen is in maintenance mode; Microsoft Agent Framework (shipped 1.0 in April 2026) is its and Semantic Kernel&#8217;s official successor — update your mental model and any bookmarked tutorials accordingly.</li>



<li>LangGraph is a low-level orchestration runtime, not a high-level agent framework — steepest learning curve, but the most control, and the choice for long-running, auditable, production-governed agents.</li>



<li>CrewAI remains the fastest path to a working multi-agent prototype and has picked up momentum as an AutoGen alternative since Microsoft&#8217;s consolidation announcement.</li>



<li>Google ADK is the most language-agnostic option (Python/TypeScript/Go/Java/Kotlin) and offers the broadest built-in menu of workflow structures — template, graph-based, dynamic, and collaborative.</li>



<li>All four now support MCP, and three of the four (Microsoft Agent Framework, Google ADK, and increasingly LangGraph via tool integrations) support the A2A protocol — cross-framework agent interoperability is becoming the norm, not the exception.</li>



<li>Pin real version floors: LangGraph 0.4+, CrewAI 0.105+, Microsoft Agent Framework 1.0+, Google ADK 2.0+ if you need graph workflows.</li>
</ul>



<h2 class="wp-block-heading">FAQs</h2>



<p class="wp-block-paragraph"><strong>Is AutoGen dead?</strong> Not dead, but deprecated in favor of Microsoft Agent Framework. It still receives bug fixes and security patches, but new feature development and Microsoft&#8217;s strategic focus have moved to the unified framework. Don&#8217;t start a new project on AutoGen in 2026.</p>



<p class="wp-block-paragraph"><strong>Can I use more than one of these frameworks together?</strong> Yes, increasingly — MCP lets any of these frameworks call the same external tools, and A2A (supported by Microsoft Agent Framework and Google ADK, with growing LangGraph support) lets agents built on different frameworks communicate directly. A common pattern is a LangGraph agent for a governed, audited core workflow calling out to an ADK-based agent for a Google-Cloud-specific task via A2A.</p>



<p class="wp-block-paragraph"><strong>Which framework is easiest for someone coming from UiPath&#8217;s low-code Agent Builder?</strong> CrewAI&#8217;s role/task/crew model maps most intuitively onto how low-code agent builders think about work — named roles with goals. LangGraph maps more closely to how you&#8217;d think about a Maestro flow or process orchestration diagram, since you&#8217;re building an explicit graph.</p>



<p class="wp-block-paragraph"><strong>Do I need to pick one framework for my whole organization?</strong> No — and given how fast this space moves, betting your entire agent strategy on one framework is risky. It&#8217;s reasonable to standardize per use case: LangGraph for governed, production-critical agents; CrewAI for internal tooling and fast prototypes; Google ADK or Microsoft Agent Framework depending on which cloud and enterprise stack a given team is already in.</p>



<p class="wp-block-paragraph"><strong>What replaced Semantic Kernel specifically, separate from AutoGen?</strong> Semantic Kernel wasn&#8217;t replaced by a different product — it was merged into Microsoft Agent Framework alongside AutoGen. If you have existing Semantic Kernel code, Microsoft&#8217;s migration guide covers moving it into the unified framework directly.</p>



<h2 class="wp-block-heading">References</h2>



<ul class="wp-block-list">
<li>LangGraph Overview — <a href="https://docs.langchain.com/oss/python/langgraph/overview" rel="nofollow noopener" target="_blank">docs.langchain.com/oss/python/langgraph/overview</a></li>



<li>LangChain Multi-Agent Patterns — <a href="https://docs.langchain.com/oss/python/langchain/multi-agent/handoffs" rel="nofollow noopener" target="_blank">docs.langchain.com/oss/python/langchain/multi-agent</a></li>



<li>LangGraph Supervisor Migration Guide — <a href="https://docs.langchain.com/oss/python/migrate/langgraph-supervisor" rel="nofollow noopener" target="_blank">docs.langchain.com/oss/python/migrate/langgraph-supervisor</a></li>



<li>Microsoft Agent Framework 1.0 announcement — <a href="https://devblogs.microsoft.com/agent-framework/microsoft-agent-framework-version-1-0/" rel="nofollow noopener" target="_blank">devblogs.microsoft.com/agent-framework</a></li>



<li>Microsoft ships production-ready Agent Framework 1.0 — <a href="https://visualstudiomagazine.com/articles/2026/04/06/microsoft-ships-production-ready-agent-framework-1-0-for-net-and-python.aspx" rel="nofollow noopener" target="_blank">Visual Studio Magazine, April 2026</a></li>



<li>Migrating Semantic Kernel / AutoGen to Microsoft Agent Framework — <a href="https://devblogs.microsoft.com/agent-framework/migrate-your-semantic-kernel-and-autogen-projects-to-microsoft-agent-framework-release-candidate/" rel="nofollow noopener" target="_blank">devblogs.microsoft.com/agent-framework</a></li>



<li>Google ADK Workflows documentation — <a href="https://adk.dev/workflows/" rel="nofollow noopener" target="_blank">adk.dev/workflows</a></li>



<li>Developer&#8217;s Guide to Multi-Agent Patterns in ADK — <a href="https://developers.googleblog.com/developers-guide-to-multi-agent-patterns-in-adk/" rel="nofollow noopener" target="_blank">Google Developers Blog</a></li>



<li>CrewAI vs LangGraph vs AutoGen comparison — <a href="https://www.datacamp.com/tutorial/crewai-vs-langgraph-vs-autogen" rel="nofollow noopener" target="_blank">DataCamp</a></li>



<li>Multi-Agent Orchestration Comparison (version guidance) — <a href="https://tensoria.fr/en/blog/multi-agent-orchestration-comparison" rel="nofollow noopener" target="_blank">Tensoria</a></li>
</ul>



<p class="wp-block-paragraph"></p>
]]></content:encoded>
					
					<wfw:commentRss>https://rpabotsworld.com/langgraph-vs-crewai-vs-microsoft-agent-framework-vs-google-adk/feed/</wfw:commentRss>
			<slash:comments>10</slash:comments>
		
		
			</item>
		<item>
		<title>The Complete Guide to Agentic AI Automation (2026)</title>
		<link>https://rpabotsworld.com/complete-guide-agentic-ai-automation/</link>
					<comments>https://rpabotsworld.com/complete-guide-agentic-ai-automation/#respond</comments>
		
		<dc:creator><![CDATA[Satish Prasad]]></dc:creator>
		<pubDate>Sun, 31 May 2026 17:52:13 +0000</pubDate>
				<category><![CDATA[Agentic AI & AI Automation]]></category>
		<category><![CDATA[agentic ai]]></category>
		<category><![CDATA[ai agents 2026]]></category>
		<category><![CDATA[ai automation]]></category>
		<category><![CDATA[crewai]]></category>
		<category><![CDATA[multi-agent systems]]></category>
		<category><![CDATA[uipath agentic]]></category>
		<guid isPermaLink="false">https://rpabotsworld.com/?p=32009</guid>

					<description><![CDATA[Everything automation professionals need to know about agentic AI in 2026 — how it works, key architectures, top frameworks, real use cases, and how it compares to traditional RPA.]]></description>
										<content:encoded><![CDATA[
<hr class="wp-block-separator has-alpha-channel-opacity"/>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph"><strong>TL;DR</strong> Agentic AI automation refers to AI systems that can perceive goals, plan multi-step strategies, use tools, and execute tasks autonomously — without being triggered step-by-step by humans. Unlike traditional RPA bots that follow fixed rules, agentic AI systems reason, adapt, and handle exceptions on their own. The market is growing at 42%+ CAGR and is already reshaping how enterprises automate knowledge work. This guide covers everything: how agents work, the five core architectures, top frameworks (CrewAI, LangGraph, AutoGen, UiPath), real industry use cases, and how to get started if you&#8217;re coming from an RPA background.</p>
</blockquote>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">1. What Is Agentic AI Automation?</h2>



<p class="wp-block-paragraph">Agentic AI automation is a category of artificial intelligence where software systems — called <strong>AI agents</strong> — are given a goal and left to figure out how to achieve it. They plan, reason, use tools, call APIs, access data, and take multi-step actions without a human directing each individual step.</p>



<p class="wp-block-paragraph">This is fundamentally different from everything that came before it.</p>



<p class="wp-block-paragraph">A traditional RPA bot follows a script. You define every click, every field, every condition. Change one screen layout and the bot breaks. A copilot or chatbot responds when you talk to it — but waits passively otherwise.</p>



<p class="wp-block-paragraph">An agentic AI system does neither. You give it an objective — &#8220;process all incoming invoice exceptions, escalate anything over $50,000, and update the ERP&#8221; — and the agent works out the steps, handles unexpected situations, and keeps going until the job is done.</p>



<p class="wp-block-paragraph">That shift — from automation that executes instructions to automation that pursues goals — is what makes agentic AI a genuinely new category, not just a smarter chatbot.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">2. Why 2026 Is the Inflection Point</h2>



<p class="wp-block-paragraph">The numbers make the trend impossible to ignore.</p>



<p class="wp-block-paragraph">The global agentic AI market hit approximately <strong>$9.89 billion in 2026</strong> and is projected to reach <strong>$57.42 billion by 2031</strong>, growing at a compound annual growth rate of over 42%. That trajectory makes it one of the fastest-growing enterprise technology segments in history.</p>



<p class="wp-block-paragraph">More telling than the market size is the adoption data. Around <strong>79% of organizations reported some level of agentic AI adoption in 2025</strong>, with 96% planning expansion in 2026. By the end of 2026, analysts expect <strong>40% of enterprise applications</strong> to embed task-specific AI agents — up from under 5% just a year earlier.</p>



<p class="wp-block-paragraph">But the most important figure for automation professionals is this: <strong>93% of US IT executives say they are extremely or very interested in agentic AI</strong>, and 37% are already using it in production, according to UiPath&#8217;s 2025 Agentic AI Report. The demand from the business side is not emerging — it is already here.</p>



<p class="wp-block-paragraph">For RPA and automation professionals, this is not a distant trend to watch. It is a skills shift happening right now, in the tools and platforms you already work with.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">3. How an AI Agent Actually Works</h2>



<p class="wp-block-paragraph">To use agentic AI effectively, you need a mental model of what is happening inside an agent at runtime. Strip away the marketing language and every agent does four things in a loop:</p>



<p class="wp-block-paragraph"><strong>Perceive → Plan → Act → Observe</strong></p>



<p class="wp-block-paragraph">Here is what each step looks like in practice:</p>



<p class="wp-block-paragraph"><strong>Perceive</strong> — The agent receives input. This might be a user instruction, a trigger from a system event, a document, an API response, or data from a previous step. The agent&#8217;s underlying language model reads and interprets this context.</p>



<p class="wp-block-paragraph"><strong>Plan</strong> — The agent reasons about what steps are needed to achieve the goal. This is where the LLM backbone earns its place. A capable model can break a complex objective into sub-tasks, sequence them logically, and identify which tools it needs.</p>



<p class="wp-block-paragraph"><strong>Act</strong> — The agent calls tools. These might be web search, database queries, API calls, file operations, code execution, or interactions with other agents. The agent takes real actions in real systems.</p>



<p class="wp-block-paragraph"><strong>Observe</strong> — The agent reads the result of its action, updates its understanding of where it stands relative to the goal, and loops back to plan the next step.</p>



<p class="wp-block-paragraph">This loop continues until the goal is achieved, a condition is met, or the agent decides it needs human input.</p>



<h3 class="wp-block-heading">The key components every agent has</h3>



<p class="wp-block-paragraph">Every production-grade AI agent relies on four building blocks:</p>



<p class="wp-block-paragraph"><strong>1. An LLM backbone</strong> — The reasoning engine. GPT-4o, Claude, Gemini, or an open-source model like Llama 3. This is what interprets instructions, plans steps, and decides what to do next.</p>



<p class="wp-block-paragraph"><strong>2. Memory</strong> — Agents use short-term memory (the current conversation context), long-term memory (vector databases like Pinecone or Chroma that persist knowledge between sessions), and episodic memory (records of what actions were taken and what happened). Without memory, every task starts from scratch.</p>



<p class="wp-block-paragraph"><strong>3. Tools</strong> — The things an agent can do. Web search, code execution, database access, API calls, file read/write, sending emails, interacting with UIs. Tools are what give an agent the ability to act in the world rather than just generate text.</p>



<p class="wp-block-paragraph"><strong>4. An orchestration layer</strong> — The runtime that manages the Perceive-Plan-Act-Observe loop, handles errors, coordinates between multiple agents if needed, and enforces governance rules. CrewAI, LangGraph, AutoGen, and UiPath&#8217;s Agent Builder all play this role.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">4. The 5 Core Agentic AI Architectures</h2>



<p class="wp-block-paragraph">Not all agents are built the same way. In 2026, five distinct patterns have emerged as the dominant architectures for enterprise agentic automation.</p>



<h3 class="wp-block-heading">Architecture 1: ReAct (Reasoning + Acting)</h3>



<p class="wp-block-paragraph">ReAct is the foundational pattern. The agent explicitly reasons before each action (&#8220;I need to find the invoice total, so I will call the ERP API&#8221;), takes an action, observes the result, and reasons again. This think-act-observe loop runs until the task is complete.</p>



<p class="wp-block-paragraph"><strong>Best for:</strong> Tasks with a clear goal but unpredictable paths. Invoice processing, document analysis, IT ticket resolution.</p>



<p class="wp-block-paragraph"><strong>Frameworks:</strong> LangChain, LangGraph, CrewAI all support ReAct agents natively.</p>



<h3 class="wp-block-heading">Architecture 2: Multi-Agent Systems (MAS)</h3>



<p class="wp-block-paragraph">Rather than one agent doing everything, multiple specialized agents collaborate. A manager agent breaks down the goal and delegates sub-tasks. Specialist agents handle research, data retrieval, writing, verification, or system calls. Results feed back up to the manager for synthesis.</p>



<p class="wp-block-paragraph"><strong>Best for:</strong> Complex workflows that benefit from parallel processing or specialized expertise. Content pipelines, financial analysis, software development workflows.</p>



<p class="wp-block-paragraph"><strong>Key insight:</strong> Multi-agent systems accounted for <strong>53.5% of enterprise agentic deployments in 2025</strong> — the dominant architecture by market share.</p>



<p class="wp-block-paragraph"><strong>Frameworks:</strong> CrewAI (crew-based delegation), AutoGen (agent conversations), LangGraph (stateful multi-agent graphs).</p>



<h3 class="wp-block-heading">Architecture 3: Agentic RAG</h3>



<p class="wp-block-paragraph">Standard RAG (Retrieval-Augmented Generation) retrieves documents once and answers a question. Agentic RAG goes further — the agent decides <em>what</em> to retrieve, <em>when</em> to retrieve it, queries multiple sources, verifies consistency between retrieved documents, and iterates if the first retrieval was insufficient.</p>



<p class="wp-block-paragraph"><strong>Best for:</strong> Knowledge-intensive work. Legal review, compliance checking, research summarization, customer query resolution from large knowledge bases.</p>



<h3 class="wp-block-heading">Architecture 4: Tool-Use / Function Calling</h3>



<p class="wp-block-paragraph">The agent is given a catalogue of tools (functions it can call) and learns to route tasks to the right tool automatically. The agent&#8217;s job is to understand a request, select the appropriate tool or sequence of tools, call them correctly, and return a result.</p>



<p class="wp-block-paragraph"><strong>Best for:</strong> Integrating AI reasoning into existing system landscapes. Connecting an LLM to your CRM, ERP, ticketing system, or data warehouse without rebuilding those systems.</p>



<p class="wp-block-paragraph"><strong>Note for RPA professionals:</strong> This is the architecture that most naturally extends existing UiPath or Power Automate workflows — the bot becomes a tool the agent can call.</p>



<h3 class="wp-block-heading">Architecture 5: Plan-and-Execute</h3>



<p class="wp-block-paragraph">The agent first creates a complete plan for the whole task, then executes each step sequentially (or in parallel). Unlike ReAct which re-reasons at every step, Plan-and-Execute is more efficient for long, predictable workflows.</p>



<p class="wp-block-paragraph"><strong>Best for:</strong> Long-horizon tasks where the overall sequence is mostly predictable, but individual steps require AI judgment. Report generation, end-to-end onboarding flows, scheduled analytical workflows.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">5. Agentic AI vs Traditional RPA: Key Differences</h2>



<p class="wp-block-paragraph">If you have spent time in RPA, the comparison table below will orient you quickly.</p>



<figure class="wp-block-table"><table class="has-fixed-layout"><thead><tr><th>Dimension</th><th>Traditional RPA</th><th>Agentic AI</th></tr></thead><tbody><tr><td>How it&#8217;s triggered</td><td>Scheduled or event-based, explicit</td><td>Goal-based, self-directed</td></tr><tr><td>What it handles</td><td>Structured, rule-based tasks</td><td>Unstructured, exception-heavy tasks</td></tr><tr><td>When it breaks</td><td>UI changes, unexpected inputs</td><td>Rarely — adapts to variation</td></tr><tr><td>Reasoning ability</td><td>None — follows a script</td><td>Full — plans and decides</td></tr><tr><td>Memory</td><td>Stateless (mostly)</td><td>Short and long-term memory</td></tr><tr><td>Tool use</td><td>Fixed integrations</td><td>Dynamic — selects tools at runtime</td></tr><tr><td>Human involvement</td><td>Defined exception routes</td><td>Human-in-the-loop as a design choice</td></tr><tr><td>Best for</td><td>High-volume, stable, predictable processes</td><td>Complex, judgement-heavy, variable processes</td></tr><tr><td>Example</td><td>Copy data from email to ERP (same fields every time)</td><td>Process an invoice exception that requires reading a contract, checking policy, and drafting a response</td></tr></tbody></table></figure>



<p class="wp-block-paragraph">The critical insight: <strong>RPA and agentic AI are not competitors</strong>. RPA handles the deterministic, high-volume execution layer. Agents handle the reasoning, exception management, and complex decision-making. The winning architecture in 2026 combines both — agents orchestrate and reason, RPA bots execute reliably in legacy systems.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">6. The Top Frameworks and Platforms</h2>



<h3 class="wp-block-heading">CrewAI</h3>



<p class="wp-block-paragraph">The most popular open-source framework for building multi-agent systems. You define agents with roles, goals, and tools, then assemble them into a &#8220;crew&#8221; that collaborates on a task. Excellent documentation and a large community make it the best starting point for practitioners. <strong>Best for:</strong> Multi-agent workflows, content pipelines, research agents. <strong>Language:</strong> Python.</p>



<h3 class="wp-block-heading">LangGraph</h3>



<p class="wp-block-paragraph">Built on top of LangChain, LangGraph lets you define agentic workflows as stateful graphs. Each node in the graph is an LLM call or tool execution. You get fine-grained control over flow, branching, and loops. More complex than CrewAI but significantly more flexible. <strong>Best for:</strong> Complex stateful workflows, enterprise-grade agents that need precise control.</p>



<h3 class="wp-block-heading">AutoGen (Microsoft)</h3>



<p class="wp-block-paragraph">AutoGen structures agentic workflows as conversations between agents. Agents exchange messages, debate approaches, and collaborate to solve problems. Particularly strong for code generation and technical workflows. <strong>Best for:</strong> Technical teams, code-focused automation, scenarios where agent debate improves output quality.</p>



<h3 class="wp-block-heading">UiPath Agent Builder</h3>



<p class="wp-block-paragraph">For RPA professionals, UiPath&#8217;s agentic automation layer is the natural entry point. Agent Builder lets you build agents that run natively in the UiPath ecosystem, access Orchestrator, call existing automations as tools, and interact with enterprise systems through existing UiPath connectors. No need to rebuild your bot library from scratch. <strong>Best for:</strong> Teams with existing UiPath investments who want to add agent-level reasoning to their automation stack.</p>



<h3 class="wp-block-heading">n8n</h3>



<p class="wp-block-paragraph">An open-source workflow automation platform that added native AI agent nodes. Lower barrier to entry than code-first frameworks, suitable for automation teams who prefer visual workflow design. <strong>Best for:</strong> Hybrid teams, no-code/low-code entry point into agentic automation.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">7. Real-World Use Cases by Industry</h2>



<h3 class="wp-block-heading">Finance and Banking</h3>



<p class="wp-block-paragraph">A bank&#8217;s accounts payable team deploys a multi-agent system to handle invoice exceptions. One agent reads incoming invoices, another cross-references the purchase order database, a third checks against the supplier contract terms, and a fourth drafts the exception response. What used to take 20 minutes per exception now takes under 2 minutes, with the agent escalating only genuine edge cases to a human reviewer.</p>



<h3 class="wp-block-heading">Healthcare</h3>



<p class="wp-block-paragraph">A hospital system uses agentic RAG to assist clinicians with patient record review. The agent retrieves relevant history, flags drug interactions against current prescriptions, summarizes recent test results, and surfaces relevant clinical guidelines — all before the physician opens the record. Human judgment is preserved for diagnosis and treatment decisions; the agent handles the information gathering.</p>



<h3 class="wp-block-heading">IT Service Management</h3>



<p class="wp-block-paragraph">An ITSM agent monitors the ticketing queue, classifies incoming tickets, searches the knowledge base for relevant solutions, attempts automated resolution for known issue types, and escalates with a full context summary when resolution fails. First-call resolution rates improve significantly because the agent never forgets a known fix.</p>



<h3 class="wp-block-heading">HR and Recruitment</h3>



<p class="wp-block-paragraph">Agentic automation handles the first stages of candidate screening. An agent reads CVs, scores them against the job specification, cross-references LinkedIn for additional context, generates a shortlist with reasoning, and drafts personalized outreach for approved candidates. Hiring managers review outputs and approve — they do not read every CV.</p>



<h3 class="wp-block-heading">Supply Chain</h3>



<p class="wp-block-paragraph">An agent monitors supplier communications, delivery confirmations, and inventory levels simultaneously. When a delivery is flagged late, the agent cross-checks the impact on production schedules, identifies alternative suppliers from the approved list, requests quotes, and surfaces a recommended action for the procurement team — all within minutes of the delay being detected.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">8. The Three Autonomy Modes Enterprises Use</h2>



<p class="wp-block-paragraph">One of the practical questions every team faces when deploying agents is how much autonomy to grant. In 2026, three operational modes have become standard:</p>



<p class="wp-block-paragraph"><strong>Assistive mode</strong> — The agent recommends actions; humans execute. The agent surfaces a suggested response, a drafted email, or a ranked list of options. The human clicks approve. Lowest risk, easiest to govern, best for high-stakes or regulated processes.</p>



<p class="wp-block-paragraph"><strong>Supervised autonomy</strong> — The agent executes but pauses for human approval at defined checkpoints. A financial agent might process transactions autonomously up to a $10,000 threshold and route anything above that for manual approval. Most enterprise deployments start here.</p>



<p class="wp-block-paragraph"><strong>Full autonomy within bounded scope</strong> — The agent runs end-to-end within tightly defined limits — specific systems, specific transaction types, explicit policy constraints. Audit logs capture every action. Reserved for well-understood, lower-risk processes with strong monitoring in place.</p>



<p class="wp-block-paragraph">The right mode depends on your risk tolerance, regulatory environment, and how well you understand the agent&#8217;s behavior in edge cases. Start narrow, measure, then expand.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">9. Governance, Safety, and Human-in-the-Loop</h2>



<p class="wp-block-paragraph">The biggest practical challenge with agentic AI in 2026 is not building agents — it is governing them. When an agent can take autonomous actions across multiple systems, you need clear answers to these questions before going to production:</p>



<p class="wp-block-paragraph"><strong>What can the agent access?</strong> Define explicit data access controls, tool permissions, and API scopes. An invoice processing agent should not have access to HR data.</p>



<p class="wp-block-paragraph"><strong>What actions require approval?</strong> Map your processes and identify which steps carry meaningful risk — financial transactions, customer communications, data deletion, external API calls with side effects. Gate these with approval workflows.</p>



<p class="wp-block-paragraph"><strong>How do you audit what happened?</strong> Every agent action should generate an audit log: what the agent was asked to do, what it decided, what tools it called, what the result was. This is non-negotiable for compliance and for debugging when things go wrong.</p>



<p class="wp-block-paragraph"><strong>What happens when the agent gets stuck?</strong> Design explicit escalation paths. An agent that cannot resolve a case within its defined parameters should hand off to a human with full context, not silently fail.</p>



<p class="wp-block-paragraph"><strong>Who owns the outcome?</strong> Governance requires accountability. Define which team owns each agent&#8217;s behavior, reviews its performance, and is responsible for its errors.</p>



<p class="wp-block-paragraph">The EU AI Act, coming into full force in 2026, classifies some automated decision-making processes as high-risk and mandates human oversight, auditability, and transparency. If your agents operate in regulated domains — finance, healthcare, HR — you need your governance framework in place before deployment, not after.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">10. How to Get Started: A Roadmap for Automation Professionals</h2>



<p class="wp-block-paragraph">If you are coming from an RPA background, you are closer to being ready for agentic AI than you might think. You already understand process analysis, exception handling, system integration, and production deployment. What you need to add is the AI reasoning layer.</p>



<p class="wp-block-paragraph"><strong>Step 1: Learn the fundamentals of LLMs</strong> — You do not need to train a model, but you need to understand how prompting works, what context windows are, what hallucinations look like, and how to write effective system prompts. The free resources from Anthropic, OpenAI, and DeepLearning.ai cover this in a few hours.</p>



<p class="wp-block-paragraph"><strong>Step 2: Build your first agent with CrewAI</strong> — Follow a hands-on tutorial and build a simple research agent. Get comfortable with the concepts of roles, tools, and task delegation before worrying about production architecture. <em>(See our tutorial: How to Build Your First AI Agent with CrewAI)</em></p>



<p class="wp-block-paragraph"><strong>Step 3: Add tools</strong> — Extend your agent with real tools: web search, file reading, an API call to a system you know. The moment your agent starts taking real actions in real systems, the model shifts from abstract concept to practical skill.</p>



<p class="wp-block-paragraph"><strong>Step 4: Connect it to your existing RPA work</strong> — If you use UiPath, explore Agent Builder and the concept of calling existing automations as agent tools. Your bot library is not obsolete — it becomes the execution layer your agents orchestrate. <em>(See: UiPath Agentic Automation: Everything You Need to Know)</em></p>



<p class="wp-block-paragraph"><strong>Step 5: Deploy something small in production</strong> — Pick a process where the stakes are low and the benefit is clear. A document summarizer, an email classifier, a first-pass ticket resolver. Get real data on how the agent performs before scaling.</p>



<p class="wp-block-paragraph"><strong>Step 6: Build your governance framework</strong> — Before any high-stakes deployment, define your autonomy mode, access controls, audit logging, and escalation paths. Governance is not a phase you do at the end — it is designed in from the start.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">11. Key Takeaways</h2>



<ul class="wp-block-list">
<li>Agentic AI automation means giving AI systems a goal and letting them reason, plan, and act to achieve it — a fundamentally different model from scripted RPA or passive copilots.</li>



<li>The market is growing at 42%+ CAGR. 40% of enterprise applications will embed AI agents by end-2026. This is not a future trend — it is happening now.</li>



<li>Every AI agent runs on a Perceive-Plan-Act-Observe loop, powered by an LLM backbone, memory, tools, and an orchestration layer.</li>



<li>Five architectures dominate in 2026: ReAct, Multi-Agent Systems, Agentic RAG, Tool-Use/Function Calling, and Plan-and-Execute. Multi-agent systems lead adoption at 53.5% market share.</li>



<li>RPA and agentic AI are complementary, not competing. Agents reason and orchestrate; bots execute reliably. Hybrid architectures are the practical path for most enterprises.</li>



<li>Top frameworks: CrewAI (multi-agent, beginner-friendly), LangGraph (flexible, stateful), AutoGen (conversation-based), UiPath Agent Builder (best for existing RPA teams).</li>



<li>Governance is the hardest part. Define autonomy modes, access controls, audit trails, and escalation paths before any production deployment.</li>



<li>RPA professionals have a head start. Process analysis, integration experience, and production mindset are exactly the skills agentic AI deployment requires.</li>
</ul>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">What to Read Next</h2>



<ul class="wp-block-list">
<li><strong><a href="https://claude.ai/rpa-to-agentic-ai-transition-guide/" rel="nofollow noopener" target="_blank">RPA to Agentic AI: The Complete Transition Guide →</a></strong> — If you&#8217;re an RPA developer, this is your next step.</li>



<li><strong><a href="https://claude.ai/build-first-ai-agent-crewai/" rel="nofollow noopener" target="_blank">How to Build Your First AI Agent with CrewAI →</a></strong> — The hands-on tutorial to get your first agent running.</li>



<li><strong><a href="https://claude.ai/uipath-agentic-automation-complete-guide/" rel="nofollow noopener" target="_blank">UiPath Agentic Automation: Everything You Need to Know →</a></strong> — For teams with existing UiPath deployments.</li>



<li><strong><a href="https://claude.ai/what-is-mcp-server-ai-agents/" rel="nofollow noopener" target="_blank">What Is an MCP Server? →</a></strong> — How the Model Context Protocol connects agents to enterprise systems.</li>
</ul>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<p class="wp-block-paragraph"><em>Written by Satish Prasad — Automation architect and founder of RPABOTS.WORLD, with 10+ years of experience in RPA, UiPath, and enterprise AI automation.</em></p>



<p class="wp-block-paragraph"><em>Last updated: June 2026. Have a correction or suggestion? <a href="https://claude.ai/contact-us/" rel="nofollow noopener" target="_blank">Contact us →</a></em></p>



<p class="wp-block-paragraph"></p>
]]></content:encoded>
					
					<wfw:commentRss>https://rpabotsworld.com/complete-guide-agentic-ai-automation/feed/</wfw:commentRss>
			<slash:comments>0</slash:comments>
		
		
			</item>
		<item>
		<title>Unlocking CrewAI Memory Types: A Guide for Technical Builders</title>
		<link>https://rpabotsworld.com/crewai-memory-types/</link>
					<comments>https://rpabotsworld.com/crewai-memory-types/#comments</comments>
		
		<dc:creator><![CDATA[Satish Prasad]]></dc:creator>
		<pubDate>Fri, 27 Jun 2025 04:10:09 +0000</pubDate>
				<category><![CDATA[Agentic AI & AI Automation]]></category>
		<category><![CDATA[crewai]]></category>
		<guid isPermaLink="false">https://rpabotsworld.com/?p=31857</guid>

					<description><![CDATA[Ever wondered how CrewAI agents remember what happened in previous conversations—or even across multiple projects? Whether you’re developing a multi-agent workflow for product automation or scaling a research assistant, CrewAI’s memory architecture is what makes your agents smarter, more consistent, and increasingly human-like over time. In this post, we’ll dive deep into CrewAI’s memory types, [&#8230;]]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Ever wondered how CrewAI agents remember what happened in previous conversations—or even across multiple projects?</p>



<p class="wp-block-paragraph">Whether you’re developing a multi-agent workflow for product automation or scaling a research assistant, CrewAI’s <strong>memory architecture</strong> is what makes your agents smarter, more consistent, and increasingly human-like over time.</p>



<p class="wp-block-paragraph">In this post, we’ll dive deep into <strong>CrewAI’s memory types</strong>, understand when and how to use them, and explore real-world use cases that demonstrate their power.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Why Memory Matters in CrewAI</h2>



<p class="wp-block-paragraph">AI agents aren&#8217;t just about processing tasks—they need <strong>context</strong>, <strong>continuity</strong>, and <strong>consistency</strong> to collaborate like humans.</p>



<p class="wp-block-paragraph">Without memory:</p>



<ul class="wp-block-list">
<li>Agents can’t build on previous knowledge.</li>



<li>Conversations reset after each interaction.</li>



<li>There’s no personalization or long-term reasoning.</li>
</ul>



<p class="wp-block-paragraph">With the right memory setup in CrewAI:</p>



<ul class="wp-block-list">
<li>Agents can pass knowledge across tasks and roles.</li>



<li>Complex workflows become manageable.</li>



<li>You get reusable intelligence baked into your systems.</li>
</ul>



<p class="wp-block-paragraph">In short, memory transforms CrewAI from a task executor into a persistent collaborator.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Types of Memory in CrewAI</h2>



<p class="wp-block-paragraph">CrewAI provides a modular memory system to suit different project needs. Let’s explore each type:</p>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f501.png" alt="🔁" class="wp-smiley" style="height: 1em; max-height: 1em;" /> 1. Short-Term Memory</h3>



<h4 class="wp-block-heading">What is it?</h4>



<p class="wp-block-paragraph">Temporary memory that lives during a single <code>Crew.kickoff()</code> session. Think of it as the &#8220;working memory&#8221; of your agents—perfect for passing data between tasks in one run.</p>



<h4 class="wp-block-heading">Use Case:</h4>



<p class="wp-block-paragraph">If Agent A generates leads and Agent B follows up on them in the same crew session, short-term memory ensures the lead data flows seamlessly.</p>



<h4 class="wp-block-heading">Benefits:</h4>



<ul class="wp-block-list">
<li>Fast, lightweight, and context-rich.</li>



<li>Enables multi-step logical reasoning.</li>



<li>Auto-cleared after session ends.</li>
</ul>



<h4 class="wp-block-heading">Example:</h4>



<pre class="wp-block-code"><code>crew = Crew(..., memory=True, short_term_memory=ShortTermMemory(...))
</code></pre>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4da.png" alt="📚" class="wp-smiley" style="height: 1em; max-height: 1em;" /> 2. Long-Term Memory</h3>



<h4 class="wp-block-heading">What is it?</h4>



<p class="wp-block-paragraph">Persistent memory that stores data across multiple sessions—allowing agents to learn over time.</p>



<h4 class="wp-block-heading">Use Case:</h4>



<p class="wp-block-paragraph">Imagine you’re running a research assistant over weeks. You want the system to remember past topics, citations, and formatting preferences. Long-term memory stores this knowledge.</p>



<h4 class="wp-block-heading">Benefits:</h4>



<ul class="wp-block-list">
<li>Historical awareness.</li>



<li>Builds institutional memory.</li>



<li>Compatible with vector stores or local DBs like SQLite.</li>
</ul>



<h4 class="wp-block-heading">Example:</h4>



<pre class="wp-block-code"><code>from crewai.memory import LongTermMemory
from crewai.memory.storage import LTMSQLiteStorage

long_term = LongTermMemory(storage=LTMSQLiteStorage(db_path="ltm.db"))

crew = Crew(..., memory=True, long_term_memory=long_term)
</code></pre>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f464.png" alt="👤" class="wp-smiley" style="height: 1em; max-height: 1em;" /> 3. Entity Memory</h3>



<h4 class="wp-block-heading">What is it?</h4>



<p class="wp-block-paragraph">Structured memory for tracking specific <strong>entities</strong>—people, tools, projects, etc.—and their evolving properties during a session.</p>



<h4 class="wp-block-heading">Use Case:</h4>



<p class="wp-block-paragraph">You’re building a CRM agent team. One agent gathers user data, another recommends products. Entity memory helps identify users and keep their preferences coherent across tasks.</p>



<h4 class="wp-block-heading">Benefits:</h4>



<ul class="wp-block-list">
<li>Structured and queryable memory.</li>



<li>Session-based consistency.</li>



<li>Especially useful in form-filling, chatbots, and RAG systems.</li>
</ul>



<h4 class="wp-block-heading">Example:</h4>



<pre class="wp-block-code"><code>from crewai.memory import EntityMemory
from crewai.memory.storage import RAGStorage

entity_memory = EntityMemory(storage=RAGStorage(...))

crew = Crew(..., memory=True, entity_memory=entity_memory)
</code></pre>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f9e0.png" alt="🧠" class="wp-smiley" style="height: 1em; max-height: 1em;" /> 4. Contextual Memory (Compositional)</h3>



<h4 class="wp-block-heading">What is it?</h4>



<p class="wp-block-paragraph">A blend of short-term, long-term, and entity memory that allows agents to maintain flow and awareness throughout task execution.</p>



<h4 class="wp-block-heading">Use Case:</h4>



<p class="wp-block-paragraph">In complex pipelines (e.g., market research → strategy creation → presentation writing), contextual memory ensures the final deliverable remains coherent and grounded.</p>



<h4 class="wp-block-heading">Benefits:</h4>



<ul class="wp-block-list">
<li>Seamless agent collaboration.</li>



<li>Maintains task flow continuity.</li>



<li>Easily enabled via <code>memory=True</code>.</li>
</ul>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f3af.png" alt="🎯" class="wp-smiley" style="height: 1em; max-height: 1em;" /> 5. User Memory (Experimental or Custom)</h3>



<h4 class="wp-block-heading">What is it?</h4>



<p class="wp-block-paragraph">Memory related to <strong>user-specific traits</strong>—like tone preference, historical queries, and interaction style.</p>



<h4 class="wp-block-heading">Use Case:</h4>



<p class="wp-block-paragraph">Personalized agents that adapt to different users over time, similar to ChatGPT’s Custom Instructions.</p>



<h4 class="wp-block-heading">Benefits:</h4>



<ul class="wp-block-list">
<li>Personalized user experiences.</li>



<li>Can be stored in external DBs or linked via ID.</li>
</ul>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Real-World Example: AI Product Team</h2>



<p class="wp-block-paragraph">Let’s say you&#8217;re building a multi-agent system to simulate a product team:</p>



<ul class="wp-block-list">
<li><strong>PM Agent</strong>: Defines requirements.</li>



<li><strong>Engineer Agent</strong>: Writes code.</li>



<li><strong>QA Agent</strong>: Tests and documents it.</li>
</ul>



<h3 class="wp-block-heading">Memory Design:</h3>



<ul class="wp-block-list">
<li><strong>Short-Term</strong>: For sharing the product spec between agents during one session.</li>



<li><strong>Entity</strong>: Tracks features (entities) and their states (in progress, tested, passed).</li>



<li><strong>Long-Term</strong>: Stores all sprint outcomes and bug reports.</li>



<li><strong>Contextual</strong>: Maintains flow from spec → code → test → report.</li>
</ul>



<p class="wp-block-paragraph">This setup makes your “crew” act like a real agile team that remembers, iterates, and improves.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Actionable Tips for Implementing Memory</h2>



<p class="wp-block-paragraph">Here’s how to get started effectively:</p>



<h3 class="wp-block-heading">1. Start With Defaults</h3>



<p class="wp-block-paragraph">Set <code>memory=True</code> in the <code>Crew()</code> config to automatically enable context management.</p>



<pre class="wp-block-preformatted">pythonCopyEdit<code>crew = Crew(..., memory=True)
</code></pre>



<h3 class="wp-block-heading">2. Choose Your Storage Wisely</h3>



<ul class="wp-block-list">
<li>Use <code>LTMSQLiteStorage</code> for long-term data.</li>



<li>For vector embeddings, plug in <code>RAGStorage</code> with your preferred backend (like Chroma or Pinecone).</li>
</ul>



<h3 class="wp-block-heading">3. Combine for Power</h3>



<p class="wp-block-paragraph">Use <strong>all three</strong> (short, long, entity) when handling:</p>



<ul class="wp-block-list">
<li>Multi-turn workflows</li>



<li>Personalization</li>



<li>Knowledge accumulation</li>
</ul>



<h3 class="wp-block-heading">4. Optimize Embedding Strategy</h3>



<p class="wp-block-paragraph">CrewAI supports custom embedders via <code>EmbeddingConfig</code>—critical for semantic memory matching.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Tools &amp; Resources</h2>



<p class="wp-block-paragraph">Here are tools and links to deepen your setup:</p>



<ul class="wp-block-list">
<li><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f517.png" alt="🔗" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <a class="" href="https://github.com/joaomdmoura/crewai" rel="nofollow noopener" target="_blank">CrewAI GitHub</a></li>



<li><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f6e0.png" alt="🛠" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <a class="" href="https://www.trychroma.com/" rel="nofollow noopener" target="_blank">ChromaDB</a> – for vector storage</li>



<li><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4d8.png" alt="📘" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <a class="" href="https://platform.openai.com/docs/guides/embeddings" rel="nofollow noopener" target="_blank">OpenAI Embeddings</a></li>



<li><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4da.png" alt="📚" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <a>LangChain Memory Docs</a></li>



<li><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4a1.png" alt="💡" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <a>CrewAI Deep Dive Guide</a></li>
</ul>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Final Thoughts: Memory Makes Agents Human-Like</h2>



<p class="wp-block-paragraph">As agent frameworks evolve, memory will define how useful and intelligent they become. CrewAI’s modular memory design empowers builders to create systems that are not just reactive—but reflective.</p>



<p class="wp-block-paragraph">So, whether you&#8217;re building a personal assistant or a multi-agent SaaS platform, don’t overlook memory.</p>



<p class="wp-block-paragraph"><strong>What type of memory have you tried in CrewAI? What challenges are you facing? Drop your thoughts below <img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f447.png" alt="👇" class="wp-smiley" style="height: 1em; max-height: 1em;" /> — let’s build smarter agents together!</strong></p>



<p class="wp-block-paragraph"></p>
]]></content:encoded>
					
					<wfw:commentRss>https://rpabotsworld.com/crewai-memory-types/feed/</wfw:commentRss>
			<slash:comments>1</slash:comments>
		
		
			</item>
		<item>
		<title>RAG vs. Agentic RAG: A Deep Dive with a CrewAI Implementation Example</title>
		<link>https://rpabotsworld.com/rag-vs-agentic-rag/</link>
					<comments>https://rpabotsworld.com/rag-vs-agentic-rag/#respond</comments>
		
		<dc:creator><![CDATA[Satish Prasad]]></dc:creator>
		<pubDate>Sat, 22 Feb 2025 05:17:04 +0000</pubDate>
				<category><![CDATA[Agentic AI & AI Automation]]></category>
		<category><![CDATA[crewai]]></category>
		<guid isPermaLink="false">https://rpabotsworld.com/?p=30277</guid>

					<description><![CDATA[Introduction Retrieval-Augmented Generation (RAG) has revolutionized how large language models (LLMs) interact with external knowledge, but as AI demands grow more complex,&#160;Agentic RAG&#160;has emerged as a transformative evolution. This blog explores the differences between RAG and Agentic RAG, their architectures, and practical implementation using&#160;CrewAI, a framework for orchestrating collaborative AI agents. By the end, you’ll [&#8230;]]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph"><strong>Introduction</strong></p>



<p class="wp-block-paragraph">Retrieval-Augmented Generation (RAG) has revolutionized how large language models (LLMs) interact with external knowledge, but as AI demands grow more complex,&nbsp;<strong>Agentic RAG</strong>&nbsp;has emerged as a transformative evolution. This blog explores the differences between RAG and Agentic RAG, their architectures, and practical implementation using&nbsp;<strong>CrewAI</strong>, a framework for orchestrating collaborative AI agents. By the end, you’ll understand how to build an Agentic RAG system that dynamically routes queries, retrieves context, and generates precise answers.</p>



<p class="wp-block-paragraph">By the end of this post, you’ll have a comprehensive understanding of Agentic RAG and its transformative potential in AI-driven systems.</p>



<h2 class="wp-block-heading"><strong>Part 1: Understanding RAG and Agentic RAG</strong></h2>



<p class="wp-block-paragraph"><strong>What is RAG?</strong></p>



<p class="wp-block-paragraph">Traditional RAG combines retrieval from external knowledge bases (e.g., vector databases) with LLM-based generation. Its workflow involves:</p>



<ol class="wp-block-list">
<li><strong>Retrieval</strong>: Fetching relevant documents using semantic search.</li>



<li><strong>Augmentation</strong>: Injecting retrieved data into the LLM’s prompt.</li>



<li><strong>Generation</strong>: Producing a response grounded in the retrieved context.</li>
</ol>



<p class="wp-block-paragraph"><strong>Limitations of RAG</strong>:</p>



<ul class="wp-block-list">
<li>Static retrieval: No iterative refinement of queries.</li>



<li>Limited adaptability: Cannot use external tools (e.g., web search, calculators).</li>



<li>No verification: Retrieved data is used &#8220;as-is&#8221; without cross-checking.</li>
</ul>



<p class="wp-block-paragraph"><strong>What is Agentic RAG?</strong></p>



<p class="wp-block-paragraph">Agentic RAG introduces&nbsp;<strong>autonomous AI agents</strong>&nbsp;to overcome RAG’s limitations. These agents:</p>



<ol class="wp-block-list">
<li><strong>Analyze and decompose queries</strong>&nbsp;into sub-tasks.</li>



<li><strong>Use tools</strong>&nbsp;(web search, APIs, calculators) to gather real-time data.</li>



<li><strong>Verify and refine</strong>&nbsp;responses iteratively.</li>
</ol>



<p class="wp-block-paragraph"><strong>Key Advantages</strong>:</p>



<ul class="wp-block-list">
<li><strong>Dynamic query optimization</strong>: Agents rephrase ambiguous queries for better retrieval.</li>



<li><strong>Multi-step reasoning</strong>: Break down complex tasks (e.g., comparing financial reports).</li>



<li><strong>Self-learning</strong>: Adapt based on user feedback.</li>
</ul>



<p class="wp-block-paragraph">Agentic RAG represents an evolution of the RAG framework by integrating intelligent agents into the retrieval and generation process. Instead of a static pipeline, Agentic RAG introduces a layer of autonomy and dynamic decision-making. Here’s what differentiates it:</p>



<ul class="wp-block-list">
<li><strong>Autonomous Agents</strong>: Specialized software agents can assess the query, decide which data sources to tap, and even decompose complex queries into smaller, manageable tasks.</li>



<li><strong>Dynamic Query Decomposition</strong>: For multifaceted queries, agents break the problem into sub-queries, execute them in parallel or sequentially, and then synthesize the results into a final coherent answer.</li>



<li><strong>Iterative Reasoning</strong>: By iterating through retrieval and generation cycles, agents can refine their results—ensuring that the final output is both accurate and contextually rich.</li>



<li><strong>Tool Integration</strong>: Agentic systems can interface with external tools (APIs, databases, custom functions) to gather additional data or perform specialized tasks, greatly expanding their capabilities.</li>
</ul>



<p class="wp-block-paragraph">This enhanced approach allows Agentic RAG systems to handle more complex, dynamic queries that require not just retrieval and generation, but also planning, reasoning, and adaptive decision-making.</p>



<h2 class="wp-block-heading"><strong>Part 2: Key Differences Between RAG and Agentic RAG</strong></h2>



<figure class="wp-block-table"><table class="has-fixed-layout"><thead><tr><th>Feature</th><th>Traditional RAG</th><th>Agentic RAG</th></tr></thead><tbody><tr><td><strong>Query Handling</strong></td><td>Single-step retrieval and generation</td><td>Multi-step reasoning with dynamic task decomposition</td></tr><tr><td><strong>Decision Making</strong></td><td>Relies on static prompt engineering</td><td>Uses agents to autonomously decide which tool or data source to use</td></tr><tr><td><strong>Adaptability</strong></td><td>Limited to pre-defined retrieval methods</td><td>Adapts in real time using routing, query planning, and tool integration</td></tr><tr><td><strong>Complex Query Support</strong></td><td>Best for straightforward Q&amp;A</td><td>Excels at complex queries, including context-aware follow-ups</td></tr><tr><td><strong>Transparency &amp; Validation</strong></td><td>Often lacks detailed source validation</td><td>Provides transparent, verifiable citations by dynamically selecting sources</td></tr></tbody></table></figure>



<p class="wp-block-paragraph">Agentic RAG’s modularity and ability to integrate multiple tools empower it to handle nuanced tasks such as generating follow-up questions, cross-referencing diverse data, and dynamically adapting the retrieval strategy—all crucial for sophisticated applications.</p>



<h2 class="wp-block-heading"><strong>Part 3:</strong> The Limitations of Traditional RAG</h2>



<p class="wp-block-paragraph">While traditional RAG is a significant step forward, it comes with several challenges:</p>



<ul class="wp-block-list">
<li><strong>Static Retrieval Processes</strong>: Traditional systems rely on a fixed retrieval strategy that may not adapt well to complex or ambiguous queries. They often lack the ability to iterate or refine the query based on intermediate results.</li>



<li><strong>Limited Multi-Step Reasoning</strong>: Without the capacity to break down a query into smaller sub-tasks, these systems can struggle with multi-faceted questions that require sequential reasoning.</li>



<li><strong>No Autonomous Decision-Making</strong>: The process is generally linear, with no mechanism to decide dynamically which external tools or additional data sources might improve the final answer.</li>



<li><strong>Inefficient Handling of Complex Tasks</strong>: When tasks involve integrating data from multiple sources or require real-time updates, traditional RAG systems may generate superficial or incomplete answers.</li>
</ul>



<p class="wp-block-paragraph">These limitations set the stage for a more advanced system—one that not only retrieves and generates but also thinks, plans, and acts. This is where Agentic RAG steps in.</p>



<figure class="wp-block-image size-large"><img fetchpriority="high" decoding="async" width="691" height="501" src="https://rpabotsworld.com/wp-content/uploads/2025/02/AI-AGENTS.drawio.svg" alt="RAG vs. Agentic RAG: A Deep Dive with a CrewAI Implementation Example 1" class="wp-image-30415" title="RAG vs. Agentic RAG: A Deep Dive with a CrewAI Implementation Example 1"></figure>



<h2 class="wp-block-heading"><strong>Part 4:</strong> Key Components of an Agentic RAG System</h2>



<p class="wp-block-paragraph">To better understand Agentic RAG, let’s break down its core components and the roles they play:</p>



<h3 class="wp-block-heading">4.1 Routing Agents</h3>



<p class="wp-block-paragraph">Routing agents serve as the first point of contact. They analyze the incoming query and decide which retrieval methods or data sources are most appropriate. For example, if a query involves code generation, the routing agent might direct the request to a specialized database of code snippets. Their primary function is to streamline the process and ensure that the right data is fetched for the given context.</p>



<h3 class="wp-block-heading">4.2 Query Planning Agents</h3>



<p class="wp-block-paragraph">Complex queries often contain multiple facets that require separate handling. Query planning agents decompose these queries into sub-queries. Each sub-query is then processed individually, and the results are later integrated into a cohesive final answer. This modular approach enhances the system’s ability to handle nuanced and multi-part questions.</p>



<h3 class="wp-block-heading">4.3 Tool Use Agents</h3>



<p class="wp-block-paragraph">Sometimes, retrieving documents alone is not enough. Tool use agents come into play by invoking external functions or APIs. For instance, if a query requires performing a mathematical calculation or fetching live data from an external API, the tool use agent will handle these additional actions. They effectively extend the system’s capabilities beyond textual data retrieval.</p>



<h3 class="wp-block-heading">4.4 ReAct Agents</h3>



<p class="wp-block-paragraph">ReAct (Reasoning and Acting) agents integrate iterative reasoning with action. They continuously refine the query based on feedback, perform necessary actions, and evaluate intermediate outputs. This iterative process allows the system to correct its course if the initial retrieval is insufficient or if new insights emerge during the process.</p>



<h3 class="wp-block-heading">4.5 Dynamic Planning and Execution Agents</h3>



<p class="wp-block-paragraph">For even more complex scenarios, dynamic planning and execution agents create a roadmap or computational graph of the tasks that need to be performed. They decide the order of operations, manage dependencies, and ensure that each step is executed optimally. This high-level planning is essential for tasks that require a sequence of actions and cannot be solved in a single pass.</p>



<p class="wp-block-paragraph">Together, these components transform a traditional RAG system into a dynamic, intelligent framework capable of handling a broad spectrum of real-world queries.</p>



<h2 class="wp-block-heading"><strong>Part 5: Building an Agentic RAG System with CrewAI</strong></h2>



<p class="wp-block-paragraph">CrewAI simplifies creating multi-agent systems where specialized agents collaborate. Let’s build a system that routes queries to a vector store (for domain-specific questions) or the web (for real-time topics).</p>



<h3 class="wp-block-heading">5.1 Define Agents </h3>



<pre class="wp-block-code"><code>router_Agent:
  role: &gt;
    Router
  goal: &gt;
    Route user question to a vectorstore or web search
  backstory: &gt;
    You are an expert at routing a user question to a vectorstore or web search .
    Use the vectorstore for questions on transformer or differential transformer.
    use web-search for question on latest news or recent topics.
    use generation for generic questions otherwise
  llm: azure/gpt-4o

retriever_Agent:
  role: &gt;
    Retriever
  goal: &gt;
    Use the information retrieved from the vectorstore to answer the question
  backstory: &gt;
    You are an assistant for question-answering tasks.
    Use the information present in the retrieved context to answer the question.
    You have to provide a clear concise answer.
  llm: azure/gpt-4o</code></pre>



<h2 class="wp-block-heading">5.2 Define Tasks</h2>



<pre class="wp-block-code"><code>router_task:
  description: &gt;
    Analyse the keywords in the question {question}"
    Based on the keywords decide whether it is eligible for a vectorstore search or a web search or generation.
    Return a single word 'vectorstore' if it is eligible for vectorstore search.
    Return a single word 'websearch' if it is eligible for web search.
    Return a single word 'generate' if it is eligible for generation.
    Do not provide any other premable or explaination.
  expected_output: &gt;
    Give a  choice 'websearch' or 'vectorstore' or 'generate' based on the question"
    Do not provide any other premable or explaination.
  agent: router_Agent

retriever_task :
  description: &gt;
    Based on the response from the router task extract information for the question {question} with the help of the respective tool.
    Use the web_serach_tool to retrieve information from the web in case the router task output is 'websearch'.
    Use the rag_tool to retrieve information from the vectorstore in case the router task output is 'vectorstore'.
    otherwise generate the output basedob your own knowledge in case the router task output is 'generate
  expected_output: &gt;
    You should analyse the output of the 'router_task'
    If the response is 'websearch' then use the web_search_tool to retrieve information from the web.
    If the response is 'vectorstore' then use the rag_tool to retrieve information from the vectorstore.
    If the response is 'generate' then use then use generation_tool .
    otherwise say i dont know if you dont know the answer
    Return a claer and consise text as response.
  agent: retriever_Agent
  </code></pre>



<h2 class="wp-block-heading">5.3 Crew.py File</h2>



<pre class="wp-block-code"><code>from crewai import Agent, Crew, Process, Task, LLM
from crewai.project import CrewBase, agent, crew, task
from crewai_tools import PDFSearchTool
from agenticrag.tools.custom_tool import GenerationTool,SearchTool
import os
from dotenv import load_dotenv

load_dotenv()

# If you want to run a snippet of code before or after the crew starts, 
# you can use the @before_kickoff and @after_kickoff decorators
# https://docs.crewai.com/concepts/crews#example-crew-class-with-decorators


config = dict(
    llm=dict(
        provider="azure_openai",
        config=dict(
            model="gpt-4o"
        ),
    ),
    embedder=dict(
        provider="azure_openai",
        config=dict(
            model="text-embedding-3-small"
        ),
    ),
)

pdf_search_tool = PDFSearchTool(config=config,pdf='my.pdf')



generation_tool=GenerationTool()
web_search_tool = SearchTool()

@CrewBase
class Agenticrag():
	"""Agenticrag crew"""

	# Learn more about YAML configuration files here:
	# Agents: https://docs.crewai.com/concepts/agents#yaml-configuration-recommended
	# Tasks: https://docs.crewai.com/concepts/tasks#yaml-configuration-recommended
	agents_config = 'config/agents.yaml'
	tasks_config = 'config/tasks.yaml'

	# If you would like to add tools to your agents, you can learn more about it here:
	# https://docs.crewai.com/concepts/agents#agent-tools
	@agent
	def router_Agent(self) -&gt; Agent:
		return Agent(
			config=self.agents_config&#91;'router_Agent'],
			verbose=True
		)

	@agent
	def retriever_Agent(self) -&gt; Agent:
		return Agent(
			config=self.agents_config&#91;'retriever_Agent'],
			verbose=True
		)

	# To learn more about structured task outputs, 
	# task dependencies, and task callbacks, check out the documentation:
	# https://docs.crewai.com/concepts/tasks#overview-of-a-task
	@task
	def router_task(self) -&gt; Task:
		return Task(
			config=self.tasks_config&#91;'router_task'],
		)

	@task
	def retriever_task (self) -&gt; Task:
		return Task(
			config=self.tasks_config&#91;'retriever_task'],
			output_file='report.md',
			tools=&#91;generation_tool,web_search_tool,pdf_search_tool]
		)

	@crew
	def crew(self) -&gt; Crew:
		"""Creates the Agenticrag crew"""
		# To learn how to add knowledge sources to your crew, check out the documentation:
		# https://docs.crewai.com/concepts/knowledge#what-is-knowledge

		return Crew(
			agents=self.agents, # Automatically created by the @agent decorator
			tasks=self.tasks, # Automatically created by the @task decorator
			process=Process.sequential,
			verbose=True,
			# process=Process.hierarchical, # In case you wanna use that instead https://docs.crewai.com/how-to/Hierarchical/
		)
</code></pre>



<h3 class="wp-block-heading">5.5 The Output </h3>



<pre class="wp-block-code"><code># Agent: Router
## Task: Analyse the keywords in the question What is AI?" Based on the keywords decide whether it is eligible for a vectorstore search or a web search or generation. Return a single word 'vectorstore' if it is eligible for vectorstore search. Return a single word 'websearch' if it is eligible for web search. Return a single word 'generate' if it is eligible for generation. Do not provide any other premable or explaination.



# Agent: Router
## Final Answer:
generate
```


# Agent: Retriever
## Task: Based on the response from the router task extract information for the question What is AI? with the help of the respective tool. Use the web_serach_tool to retrieve information from the web in case the router task output is 'websearch'. Use the rag_tool to retrieve information from the vectorstore in case the router task output is 'vectorstore'. otherwise generate the output basedob your own knowledge in case the router task output is 'generate



# Agent: Retriever
## Thought: Thought: Based on the router task output, which is "generate", I will use the generation_tool to answer the question "What is AI?"
## Using tool: Generation_tool
## Tool Input:
"{\"query\": \"What is AI?\"}"
## Tool Output:
content='AI, or **Artificial Intelligence**, refers to the simulation of human intelligence in machines that are programmed to think, learn, and make decisions like humans. These systems are designed to perform tasks that typically require human intelligence, such as problem-solving, understanding natural language, recognizing patterns, and adapting to new information.\n\n### Key Components of AI:\n1. **Machine Learning (ML):**\n   - A subset of AI that enables machines to learn from data and improve their performance over time without being explicitly programmed.\n   - Example: A recommendation system on Netflix or Amazon.\n\n2. **Natural Language Processing (NLP):**\n   - The ability of machines to understand, interpret, and respond to human language.\n   - Example: Virtual assistants like Siri, Alexa, or chatbots.\n\n3. **Computer Vision:**\n   - The ability of machines to interpret and analyze visual data, such as images or videos.\n   - Example: Facial recognition or object detection.\n\n4. **Robotics:**\n   - The use of AI to control robots that can perform tasks autonomously or semi-autonomously.\n   - Example: Self-driving cars or robotic arms in manufacturing.\n\n5. **Deep Learning:**\n   - A more advanced subset of machine learning that uses neural networks to mimic the way the human brain processes information.\n   - Example: Image recognition or voice synthesis.\n\n### Types of AI:\n1. **Narrow AI (Weak AI):**\n   - AI systems designed to perform a specific task or a narrow range of tasks.\n   - Example: Spam filters, chess-playing programs.\n\n2. **General AI (Strong AI):**\n   - Hypothetical AI that can perform any intellectual task a human can do, with the ability to reason, learn, and adapt across a wide range of activities.\n   - Example: This level of AI does not yet exist.\n\n3. **Superintelligent AI:**\n   - A theoretical AI that surpasses human intelligence in all aspects, including creativity, problem-solving, and decision-making.\n   - Example: A concept often explored in science fiction.\n\n### Applications of AI:\n- **Healthcare:** Diagnosing diseases, drug discovery, and personalized medicine.\n- **Finance:** Fraud detection, algorithmic trading, and credit scoring.\n- **Transportation:** Autonomous vehicles and traffic management.\n- **Entertainment:** Content recommendations and video game AI.\n- **Customer Service:** Chatbots and virtual assistants.\n- **Manufacturing:** Predictive maintenance and quality control.\n\n### Benefits of AI:\n- Increased efficiency and productivity.\n- Automation of repetitive tasks.\n- Enhanced decision-making through data analysis.\n- Improved accuracy in various fields, such as medicine and engineering.\n\n### Challenges and Concerns:\n- Ethical issues, such as bias in AI algorithms.\n- Job displacement due to automation.\n- Privacy concerns with data collection and surveillance.\n- The potential risks of creating highly autonomous systems.\n\nIn summary, AI is a transformative technology with the potential to revolutionize industries and improve lives, but it also requires careful consideration of its ethical and societal implications.' additional_kwargs={'refusal': None} response_metadata={'token_usage': {'completion_tokens': 610, 'prompt_tokens': 11, 'total_tokens': 621, 'completion_tokens_details': {'accepted_prediction_tokens': 0, 'audio_tokens': 0, 'reasoning_tokens': 0, 'rejected_prediction_tokens': 0}, 'prompt_tokens_details': {'audio_tokens': 0, 'cached_tokens': 0}}, 'model_name': 'gpt-4o-2024-11-20', 'system_fingerprint': 'fp_b705f0c291', 'prompt_filter_results': &#91;{'prompt_index': 0, 'content_filter_results': {'hate': {'filtered': False, 'severity': 'safe'}, 'jailbreak': {'filtered': False, 'detected': False}, 'self_harm': {'filtered': False, 'severity': 'safe'}, 'sexual': {'filtered': False, 'severity': 'safe'}, 'violence': {'filtered': False, 'severity': 'safe'}}}], 'finish_reason': 'stop', 'logprobs': None, 'content_filter_results': {'hate': {'filtered': False, 'severity': 'safe'}, 'protected_material_code': {'filtered': False, 'detected': False}, 'protected_material_text': {'filtered': False, 'detected': False}, 'self_harm': {'filtered': False, 'severity': 'safe'}, 'sexual': {'filtered': False, 'severity': 'safe'}, 'violence': {'filtered': False, 'severity': 'safe'}}} id='run-fed56c6e-d50e-48d7-b1d4-f50db5be4a1e-0' usage_metadata={'input_tokens': 11, 'output_tokens': 610, 'total_tokens': 621, 'input_token_details': {'audio': 0, 'cache_read': 0}, 'output_token_details': {'audio': 0, 'reasoning': 0}}


# Agent: Retriever
## Final Answer:
AI, or **Artificial Intelligence**, refers to the simulation of human intelligence in machines that are programmed to think, learn, and make decisions like humans. These systems are designed to perform tasks that typically require human intelligence, such as problem-solving, understanding natural language, recognizing patterns, and adapting to new information.

### Key Components of AI:
1. **Machine Learning (ML):**
   - A subset of AI that enables machines to learn from data and improve their performance over time without being explicitly programmed.
   - Example: A recommendation system on Netflix or Amazon.

2. **Natural Language Processing (NLP):**
   - The ability of machines to understand, interpret, and respond to human language.
   - Example: Virtual assistants like Siri, Alexa, or chatbots.

3. **Computer Vision:**
   - The ability of machines to interpret and analyze visual data, such as images or videos.
   - Example: Facial recognition or object detection.

4. **Robotics:**
   - The use of AI to control robots that can perform tasks autonomously or semi-autonomously.
   - Example: Self-driving cars or robotic arms in manufacturing.</code></pre>



<p class="wp-block-paragraph"></p>



<h2 class="wp-block-heading"><strong>Part 6:</strong> Challenges in Deploying Agentic RAG</h2>



<p class="wp-block-paragraph">While the benefits are substantial, implementing Agentic RAG is not without its challenges:</p>



<ul class="wp-block-list">
<li><strong>Data Quality and Consistency</strong>: The system’s performance is highly dependent on the quality and consistency of the underlying data. Inconsistent or outdated data can lead to inaccuracies.</li>



<li><strong>Integration Complexity</strong>: Seamlessly integrating multiple agents, tools, and external data sources requires careful design and robust infrastructure.</li>



<li><strong>Computational Resources</strong>: Multi-step reasoning and dynamic retrieval can be resource-intensive, especially when processing real-time data or deploying at scale.</li>



<li><strong>Ethical and Bias Considerations</strong>: As with all AI systems, ensuring fairness and mitigating biases in the training data are critical to maintaining trust.</li>



<li><strong>Ongoing Maintenance</strong>: Agentic RAG systems require continuous updates and maintenance to stay current with new data and evolving user needs.</li>
</ul>



<p class="wp-block-paragraph">Addressing these challenges involves adopting best practices in data management, investing in scalable infrastructure, and implementing robust feedback loops to refine the system over time.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">The Future of Agentic RAG</h2>



<p class="wp-block-paragraph">The evolution of Agentic RAG is poised to redefine how we interact with information. Emerging trends indicate several exciting directions:</p>



<ul class="wp-block-list">
<li><strong>Multimodal Capabilities</strong>: Future systems may integrate text, images, audio, and video data, enabling richer context and more immersive user experiences.</li>



<li><strong>Personalization</strong>: Leveraging user profiles and interaction histories, Agentic RAG can deliver hyper-personalized responses tailored to individual needs.</li>



<li><strong>Enhanced Explainability</strong>: As users demand more transparency, future systems will provide clearer explanations of how responses are generated, building trust and accountability.</li>



<li><strong>Integration with Edge Computing</strong>: Deploying Agentic RAG models closer to the data source (e.g., on mobile devices or local servers) can reduce latency and improve responsiveness.</li>



<li><strong>Industry-Specific Solutions</strong>: Customized Agentic RAG applications for sectors like healthcare, finance, and legal services will become increasingly prevalent, offering specialized insights and support.</li>
</ul>



<p class="wp-block-paragraph">These advancements will further blur the lines between information retrieval, decision-making, and autonomous action, paving the way for AI systems that are not only intelligent but also deeply integrated into our everyday workflows.</p>



<h2 class="wp-block-heading">Further Reading and Resources</h2>



<ul class="wp-block-list">
<li><strong>Analytics Vidhya’s Comprehensive Guide</strong>: For a deeper dive into the differences between traditional RAG and Agentic RAG, explore detailed comparisons and technical insights.<br><a href="https://www.analyticsvidhya.com/blog/2024/11/rag-vs-agentic-rag/" target="_blank" rel="noreferrer noopener nofollow">analyticsvidhya.com</a></li>



<li><strong>Aisera’s Blog on Agentic RAG</strong>: Gain additional context on the evolution of Agentic RAG and its real-world applications.<br><a href="https://aisera.com/blog/agentic-rag/" target="_blank" rel="noreferrer noopener nofollow">aisera.com</a></li>



<li><strong>CrewAI Implementations on GitHub</strong>: Check out open-source implementations of Agentic RAG workflows using CrewAI to see practical code examples.<br><a href="https://github.com/pavanbelagatti/Agentic-RAG-LangChain-CrewAI/blob/main/crew-agentic-pav.ipynb" target="_blank" rel="noreferrer noopener nofollow">github.com</a></li>



<li><strong>AWS Machine Learning Blog</strong>: Learn how leading platforms like AWS are deploying agentic AI solutions in production environments.<br><a href="https://aws.amazon.com/blogs/machine-learning/build-agentic-ai-solutions-with-deepseek-r1-crewai-and-amazon-sagemaker-ai/" target="_blank" rel="noreferrer noopener nofollow">aws.amazon.com</a></li>
</ul>



<p class="wp-block-paragraph"></p>
]]></content:encoded>
					
					<wfw:commentRss>https://rpabotsworld.com/rag-vs-agentic-rag/feed/</wfw:commentRss>
			<slash:comments>0</slash:comments>
		
		
			</item>
		<item>
		<title>Unlocking CrewAI Knowledge Feature: A Practical Guide with Examples</title>
		<link>https://rpabotsworld.com/crewai-knowledge-feature/</link>
					<comments>https://rpabotsworld.com/crewai-knowledge-feature/#respond</comments>
		
		<dc:creator><![CDATA[Satish Prasad]]></dc:creator>
		<pubDate>Wed, 19 Feb 2025 07:33:02 +0000</pubDate>
				<category><![CDATA[Agentic AI & AI Automation]]></category>
		<category><![CDATA[crewai]]></category>
		<guid isPermaLink="false">https://rpabotsworld.com/?p=30336</guid>

					<description><![CDATA[CrewAI’s&#160;Knowledge&#160;system is a game-changer for developers and businesses looking to enhance AI agents with contextual, domain-specific data. This blog dives into how to leverage this feature effectively, complete with real-world examples and actionable insights. What is Knowledge in CrewAI? The Knowledge system allows AI agents to access and utilize external data sources—like PDFs, CSVs, or APIs—during task [&#8230;]]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">CrewAI’s&nbsp;<strong>Knowledge</strong>&nbsp;system is a game-changer for developers and businesses looking to enhance AI agents with contextual, domain-specific data. This blog dives into how to leverage this feature effectively, complete with real-world examples and actionable insights.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">What is Knowledge in CrewAI?</h2>



<p class="wp-block-paragraph">The <strong>Knowledge</strong> system allows AI agents to access and utilize external data sources—like PDFs, CSVs, or APIs—during task execution. Think of it as equipping your agents with a dynamic reference library, enabling them to ground responses in factual information and improve decision-making .</p>



<p class="wp-block-paragraph"><strong>Key Benefits</strong>:</p>



<ul class="wp-block-list">
<li><strong>Domain-Specific Expertise</strong>: Agents can access specialized data (e.g., product manuals, financial reports) .</li>



<li><strong>Real-Time Context</strong>: Maintain continuity across interactions, such as customer support conversations .</li>



<li><strong>Flexibility</strong>: Supports structured (CSV, JSON) and unstructured (PDF, text) data .</li>
</ul>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Supported Knowledge Sources</h2>



<p class="wp-block-paragraph"><a href="https://rpabotsworld.com/memory-in-ai-agents/" data-type="post" data-id="30206">CrewAI supports </a>a wide range of knowledge sources, which can be broadly categorized as follows:</p>



<ul class="wp-block-list">
<li><strong>Text Sources:</strong> Raw strings, text files, and PDFs.</li>



<li><strong>Structured Data:</strong> CSV, Excel, and JSON documents.</li>



<li><strong>Custom Sources:</strong> Easily extendable to incorporate APIs or any other data by inheriting from the base knowledge source class.</li>
</ul>



<p class="wp-block-paragraph">This versatility means you can choose the right type of content for your agents’ tasks, whether you’re building a support agent or a research assistant.</p>



<h2 class="wp-block-heading">Setting Up Knowledge Sources</h2>



<h3 class="wp-block-heading">Basic Configuration</h3>



<ol start="1" class="wp-block-list">
<li><strong>Folder Structure</strong>: Create a <code>knowledge</code> directory in your project root and place files there (e.g., <code>knowledge/report.pdf</code>) .</li>



<li><strong>Define Sources</strong>: Use built-in classes like <code>PDFKnowledgeSource</code> or <code>CSVKnowledgeSource</code> to load documents.</li>
</ol>



<pre class="wp-block-code"><code>from crewai.knowledge.source.pdf_knowledge_source import PDFKnowledgeSource

# Load a PDF from the knowledge directory
pdf_source = PDFKnowledgeSource(
    file_path="report.pdf",  # Relative to the knowledge folder
    chunk_size=4000,         # Split into 4000-character chunks
    chunk_overlap=200        # Overlap chunks for context retention
)

# Add to your Crew
crew = Crew(
    agents=&#91;researcher, writer],
    tasks=&#91;task],
    knowledge_sources=&#91;pdf_source]
)</code></pre>



<p class="wp-block-paragraph"><em>Note</em>: If you encounter metadata errors (e.g., <code>Expected metadata to be a non-empty dict</code>), add dummy metadata like <code>metadata={"title": "dummy"}</code> .</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading">Advanced Configuration</h2>



<h3 class="wp-block-heading">1. Chunking &amp; Embeddings</h3>



<ul class="wp-block-list">
<li><strong>Chunking</strong>: Adjust <code>chunk_size</code> and <code>chunk_overlap</code> to balance context retention and processing efficiency 12.</li>



<li><strong>Embeddings</strong>: Use providers like Google (<code>text-embedding-004</code>) or OpenAI for vector storage.</li>
</ul>



<p class="wp-block-paragraph"><strong>Example: Custom Embeddings</strong></p>



<pre class="wp-block-code"><code>crew = Crew(
    ...
    embedder={
        "provider": "google",
        "config": {"model": "text-embedding-004", "api_key": "YOUR_KEY"}
    }
)</code></pre>



<h3 class="wp-block-heading">Custom Knowledge Sources</h3>



<p class="wp-block-paragraph">Extend&nbsp;<code>BaseKnowledgeSource</code>&nbsp;to integrate real-time data.</p>



<p class="wp-block-paragraph"><strong>Example: Space News API Integration</strong></p>



<pre class="wp-block-code"><code>from crewai.knowledge.source.base_knowledge_source import BaseKnowledgeSource
import requests

class SpaceNewsKnowledgeSource(BaseKnowledgeSource):
    def load_content(self):
        response = requests.get("https://api.spaceflightnewsapi.net/v4/articles")
        articles = response.json()&#91;"results"]
        return self._format_articles(articles)
    
    def _format_articles(self, articles):
        return "\n".join(&#91;f"{article&#91;'title']}: {article&#91;'summary']}" for article in articles])

# Assign to an agent
agent = Agent(
    role="Space News Analyst",
    knowledge_sources=&#91;SpaceNewsKnowledgeSource()]
)</code></pre>



<h2 class="wp-block-heading">Quickstart Example: Using a String-Based Knowledge Source</h2>



<p class="wp-block-paragraph">Let’s start with a simple example. Imagine you have a snippet of text about a user, and you want your agent to answer questions using that information. The following code demonstrates how to set up a string-based knowledge source:</p>



<pre class="wp-block-code"><code>from crewai import Agent, Task, Crew, Process, LLM
from crewai.knowledge.source.string_knowledge_source import StringKnowledgeSource

# Create a knowledge source with user data
content = "Users name is John. He is 30 years old and lives in San Francisco."
string_source = StringKnowledgeSource(content=content)

# Initialize an LLM with a deterministic setting
llm = LLM(model="gpt-4o-mini", temperature=0)

# Create an agent that leverages this knowledge
agent = Agent(
    role="About User",
    goal="You know everything about the user.",
    backstory="You are a master at understanding people and their preferences.",
    verbose=True,
    allow_delegation=False,
    llm=llm,
)

# Define a task where the agent answers a user question
task = Task(
    description="Answer the following questions about the user: {question}",
    expected_output="An answer to the question.",
    agent=agent,
)

# Create a crew and attach the knowledge source
crew = Crew(
    agents=&#91;agent],
    tasks=&#91;task],
    verbose=True,
    process=Process.sequential,
    knowledge_sources=&#91;string_source],
)

# Kick off the crew with a specific question
result = crew.kickoff(inputs={"question": "What city does John live in and how old is he?"})
</code></pre>



<h2 class="wp-block-heading">Expanding Your Horizons: File-Based Knowledge Sources</h2>



<p class="wp-block-paragraph">Beyond raw strings, CrewAI supports various file formats to suit different data needs:</p>



<ul class="wp-block-list">
<li><strong>Text Files:</strong> Use the <code>TextFileKnowledgeSource</code> to load data from <code>.txt</code> files.</li>



<li><strong>PDFs:</strong> The <code>PDFKnowledgeSource</code> helps your agent extract information from PDF documents.</li>



<li><strong>CSV, Excel, and JSON:</strong> Use their respective knowledge sources to integrate structured data seamlessly.</li>
</ul>



<p class="wp-block-paragraph">For instance, if you want to extract information from a CSV file containing product details, simply instantiate the <code>CSVKnowledgeSource</code> with the path to your file and add it to your crew’s knowledge sources.</p>



<h2 class="wp-block-heading">Custom Knowledge Source:  PDF Source Example</h2>



<pre class="wp-block-code"><code>from crewai import Agent, Crew, Process, Task
from crewai.project import CrewBase, agent, crew, task
from crewai.knowledge.source.pdf_knowledge_source import PDFKnowledgeSource

# Initialize the PDF knowledge source with a file path
pdf_source = PDFKnowledgeSource(
    file_paths=&#91;"meta_quest_manual.pdf"]
)

@CrewBase
class MetaQuestKnowledge():
    """MetaQuestKnowledge crew"""

    # Configurations for agents and tasks are stored in external YAML files.
    agents_config = 'config/agents.yaml'
    tasks_config = 'config/tasks.yaml'

    @agent
    def meta_quest_expert(self) -> Agent:
        # Create an agent using the configuration for the meta quest expert.
        # The agent will leverage the PDF knowledge source during tasks.
        return Agent(
            config=self.agents_config&#91;'meta_quest_expert'],
            verbose=True
        )

    @task
    def answer_question_task(self) -> Task:
        # Define a task that is responsible for answering user questions.
        # Task details are provided in the YAML configuration.
        return Task(
            config=self.tasks_config&#91;'answer_question_task'],
        )

    @crew
    def crew(self) -> Crew:
        """Creates the MetaQuestKnowledge crew"""
        # Assemble the crew by collecting all agents and tasks.
        # The PDF knowledge source is added to allow agents to use the content
        # of the PDF when processing queries.
        return Crew(
            agents=self.agents,  # Automatically populated by the @agent decorator
            tasks=self.tasks,    # Automatically populated by the @task decorator
            process=Process.sequential,
            verbose=True,
            knowledge_sources=&#91;
                pdf_source
            ]
        )
</code></pre>



<h2 class="wp-block-heading">Final Thoughts</h2>



<p class="wp-block-paragraph">By integrating a PDF knowledge source, you empower your AI agents with the ability to extract and use real-world data from documents. This example illustrates how to set up a clean, modular Crew using CrewAI—leveraging external configurations and the power of a PDF knowledge source. Whether you&#8217;re building a support bot, a research assistant, or any task-specific agent, this approach ensures that your agents remain well-informed and contextually accurate.</p>



<p class="wp-block-paragraph">Happy coding, and may your AI projects be ever more knowledgeable!</p>



<p class="wp-block-paragraph"></p>
]]></content:encoded>
					
					<wfw:commentRss>https://rpabotsworld.com/crewai-knowledge-feature/feed/</wfw:commentRss>
			<slash:comments>0</slash:comments>
		
		
		<media:thumbnail url="https://rpabotsworld.com/wp-content/uploads/2023/05/Unleashing-the-Power-of-Low-Code.jpg" />	</item>
		<item>
		<title>Memory in AI Agents: Unlocking Contextual Intelligence with CrewAI and AutoGen</title>
		<link>https://rpabotsworld.com/memory-in-ai-agents/</link>
					<comments>https://rpabotsworld.com/memory-in-ai-agents/#comments</comments>
		
		<dc:creator><![CDATA[Satish Prasad]]></dc:creator>
		<pubDate>Fri, 14 Feb 2025 06:19:27 +0000</pubDate>
				<category><![CDATA[Agentic AI & AI Automation]]></category>
		<category><![CDATA[crewai]]></category>
		<guid isPermaLink="false">https://rpabotsworld.com/?p=30206</guid>

					<description><![CDATA[Understanding AI Agent Memory: A Human Analogy Artificial Intelligence (AI) agents, like human brains, need memory to function effectively. Just as humans rely on different types of memory to process, retain, and recall information, AI-powered Agent Builders implement sophisticated memory architectures to enhance decision-making and user interactions. The Three Layers of Memory in AI Agents [&#8230;]]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph"><strong>Understanding AI Agent Memory: A Human Analogy</strong></p>



<p class="wp-block-paragraph">Artificial Intelligence (AI) agents, like human brains, need memory to function effectively. Just as humans rely on different types of memory to process, retain, and recall information, AI-powered Agent Builders implement sophisticated memory architectures to enhance decision-making and user interactions.</p>



<h2 class="wp-block-heading">The Three Layers of Memory in AI Agents (Like the Human Brain)</h2>



<h3 class="wp-block-heading">1. <strong>Sensory Memory – Like Human Senses</strong></h3>



<p class="wp-block-paragraph">Before processing any information, humans first perceive their surroundings through sight, sound, touch, and other senses. This fleeting sensory input is either ignored or passed to short-term memory if it is deemed important.</p>



<p class="wp-block-paragraph">In AI Agent Builders, <strong>sensory memory</strong> functions similarly, capturing raw inputs from text, voice, images, and sensor data. This stage is crucial for quick responsiveness, allowing AI to react in real time.</p>



<h3 class="wp-block-heading">2. <strong>Short-Term Memory – Like Working Memory</strong></h3>



<p class="wp-block-paragraph">Humans can only hold a small amount of information in their working memory at any given time. This enables them to focus on relevant details while performing a task, like remembering a phone number long enough to dial it.</p>



<p class="wp-block-paragraph">AI agents use <strong>short-term memory</strong> to temporarily store recent interactions, such as ongoing conversations in a chatbot or the last command given to a voice assistant. This memory fades quickly unless it’s reinforced or saved for long-term use.</p>



<h3 class="wp-block-heading">3. <strong>Long-Term Memory – Like Stored Knowledge</strong></h3>



<p class="wp-block-paragraph">Humans retain past experiences, lessons, and learned concepts in long-term memory, which allows them to recall past events, recognize patterns, and apply knowledge to new situations.</p>



<p class="wp-block-paragraph">Similarly, <strong>long-term memory in AI agents</strong> enables them to store historical data, user preferences, and behavioral patterns for future interactions. This ensures AI systems can deliver personalized recommendations, recall past user queries, and improve their responses over time.</p>



<figure class="wp-block-image size-full is-resized"><img decoding="async" width="100" height="272" src="https://rpabotsworld.com/wp-content/uploads/2025/02/4caeb05f-704e-49de-88f4-41d3fe05605a.svg" alt="Memory in AI Agents: Unlocking Contextual Intelligence with CrewAI and AutoGen 2" class="wp-image-30210" style="width:770px;height:auto" title="Memory in AI Agents: Unlocking Contextual Intelligence with CrewAI and AutoGen 2"></figure>



<h2 class="wp-block-heading">Techniques for Building Memory (Inspired by Human Learning)</h2>



<p class="wp-block-paragraph">To mimic human cognitive abilities, AI Agent Builders employ several advanced memory techniques:</p>



<ul class="wp-block-list">
<li><strong>Memory Networks</strong> – Just like neural pathways in the brain, these help AI agents retrieve and process stored information efficiently.</li>



<li><strong>Recurrent Neural Networks (RNNs) &amp; LSTMs</strong> – These models function similarly to how humans remember sequences, such as learning a new language or recalling past conversations.</li>



<li><strong>Transformers</strong> – Analogous to rapid recall and focused attention in the human brain, these models allow AI to process large amounts of text and context efficiently.</li>



<li><strong>Knowledge Graphs</strong> – Similar to how humans associate concepts and ideas, AI uses interconnected knowledge structures to enhance decision-making.</li>



<li><strong>Episodic Memory Systems</strong> – Like humans recalling personal experiences, AI can remember specific user interactions to provide more contextual responses.</li>
</ul>



<h2 class="wp-block-heading">Managing Forgetting &amp; Privacy (Like Human Forgetting)</h2>



<p class="wp-block-paragraph">Forgetting is just as crucial as remembering. Humans don’t store every single detail of their lives—our brains prioritize important memories and let go of irrelevant ones. AI must do the same to ensure efficiency, relevance, and privacy:</p>



<ul class="wp-block-list">
<li><strong>Time Decay</strong> – Just like how memories fade over time, AI discards older, less relevant data automatically.</li>



<li><strong>Relevance Scoring</strong> – AI prioritizes important information, similar to how humans remember significant events over trivial details.</li>



<li><strong>User-Controlled Deletion</strong> – Like humans selectively forgetting things, AI should allow users to erase specific data for privacy and compliance.</li>
</ul>



<p class="wp-block-paragraph"></p>



<h3 class="wp-block-heading">Understanding Memory in AI Agent Builders: CrewAI vs. AutoGen</h3>



<p class="wp-block-paragraph">Memory is a critical component in<a href="https://rpabotsworld.com/working-with-crewai/" data-type="link" data-id="https://rpabotsworld.com/working-with-crewai/"> AI agents</a>, enabling them to retain context, learn from interactions, and deliver coherent, personalized responses. Below, we explore how memory is implemented in <strong>CrewAI</strong> and <strong>AutoGen</strong>, two popular frameworks for building AI agents, with examples and key insights from their architectures.</p>



<h3 class="wp-block-heading">What is Memory in Agent Builders?</h3>



<p class="wp-block-paragraph">Memory systems in agent frameworks typically include:</p>



<ul class="wp-block-list">
<li><strong>Short-Term Memory:</strong> Temporarily holds recent interactions and context for the current task.</li>



<li><strong>Long-Term Memory:</strong> Stores valuable insights, learnings, and past interactions for future reference.</li>



<li><strong>Entity Memory:</strong> Captures specific details about entities (like users, products, or concepts) encountered during tasks.</li>



<li><strong>Contextual Memory:</strong> Integrates various memory types to maintain overall context over a series of interactions.</li>
</ul>



<p class="wp-block-paragraph">This structured memory allows agents to “remember” what happened before, adapt to new information, and generate more coherent, context-aware responses.</p>



<p class="wp-block-paragraph">These memory types allow agents to:</p>



<ul class="wp-block-list">
<li>Maintain conversational coherence.</li>



<li>Personalize responses based on past interactions.</li>



<li>Optimize task execution through adaptive learning</li>
</ul>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f9e0.png" alt="🧠" class="wp-smiley" style="height: 1em; max-height: 1em;" /> Example: Memory in CrewAI</h3>



<p class="wp-block-paragraph"><a href="https://rpabotsworld.com/working-with-crewai/" data-type="link" data-id="https://rpabotsworld.com/working-with-crewai/">CrewAI features</a> a sophisticated memory system:</p>



<ul class="wp-block-list">
<li><strong>Short-Term Memory:</strong> Uses a RAG (Retrieval-Augmented Generation) approach with a vector store (e.g., Chroma) to temporarily store recent interactions.</li>



<li><strong>Long-Term Memory:</strong> Persists high-value insights using a SQLite storage backend.</li>



<li><strong>Entity Memory:</strong> Organizes information about key entities encountered during tasks.</li>
</ul>



<p class="wp-block-paragraph"><strong>Key Features</strong>:</p>



<ul class="wp-block-list">
<li><strong>Comprehensive Memory System</strong>: Integrates short-term, long-term, and entity memory, similar to LangGraph .</li>



<li><strong>Contextual Awareness</strong>: Agents retain session-specific data (e.g., user preferences or task progress) .</li>



<li><strong>Customizable Storage</strong>: Uses external databases (e.g., vector stores) for efficient retrieval </li>
</ul>



<pre class="wp-block-code"><code>from crewai import Crew, Process
from crewai.memory import LongTermMemory, ShortTermMemory, EntityMemory
from crewai.memory.storage import LTMSQLiteStorage, RAGStorage

crew = Crew(
    agents=&#91;...],
    tasks=&#91;...],
    process=Process.sequential,
    memory=True,
    long_term_memory=LongTermMemory(
        storage=LTMSQLiteStorage(db_path="./long_term_memory.db")
    ),
    short_term_memory=ShortTermMemory(
        storage=RAGStorage(
            embedder_config={"provider": "openai", "config": {"model": "text-embedding-3-small"}},
            type="short_term",
            path="./short_term_memory"
        )
    ),
    entity_memory=EntityMemory(
        storage=RAGStorage(
            embedder_config={"provider": "openai", "config": {"model": "text-embedding-3-small"}},
            type="short_term",
            path="./entity_memory"
        )
    ),
    verbose=True,
)
</code></pre>



<p class="wp-block-paragraph">In this example, CrewAI’s memory system ensures that agents have both immediate context (via short-term memory) and accumulated insights (via long-term and entity memory) to inform their decisions.</p>



<p class="wp-block-paragraph"><strong>Use Case</strong>:</p>



<ul class="wp-block-list">
<li>A customer support team where agents remember user issues across interactions to avoid repetition</li>
</ul>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f9e0.png" alt="🧠" class="wp-smiley" style="height: 1em; max-height: 1em;" /> Example: Memory in AutoGen</h3>



<p class="wp-block-paragraph">AutoGen focuses on <strong>conversational workflows</strong> and <strong>code execution</strong>, with memory primarily tied to chat history and task context.</p>



<p class="wp-block-paragraph">AutoGen also supports memory to help conversational agents maintain context across interactions:</p>



<ul class="wp-block-list">
<li><strong>Long-Term Memory Support:</strong> AutoGen can integrate with advanced memory management platforms like Zep and Mem0. This allows agents to remember conversation history and learn from it.</li>



<li><strong>Conversational Memory:</strong> Since AutoGen treats workflows as conversations, it maintains context by continuously updating the conversation history. This lets agents refer back to earlier parts of the discussion.</li>
</ul>



<p class="wp-block-paragraph"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1fae1.png" alt="🫡" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <strong>Key Features</strong>:</p>



<ul class="wp-block-list">
<li><strong>Conversation-Driven Memory</strong>: Agents retain chat histories to maintain context in multi-step interactions .</li>



<li><strong>Human-in-the-Loop</strong>: Allows manual intervention to adjust memory (e.g., correcting errors mid-task) .</li>



<li><strong>Modular Design</strong>: Supports custom memory extensions via code execution (e.g., storing outputs in databases)</li>
</ul>



<pre class="wp-block-code"><code>from autogen import AgentBuilder, ConversationManager

# Build an agent with memory support
agent = AgentBuilder(
    name="ResearchAgent",
    role="Conducts research and retains context from prior queries",
    memory_enabled=True,  # Enable memory feature
    memory_provider="zep"  # Use an external provider like Zep for long-term memory
)

conversation = ConversationManager(agent=agent)
response = conversation.send("What are the latest trends in AI?")
print(response)
</code></pre>



<p class="wp-block-paragraph">Here, AutoGen’s integration with memory providers (e.g., Zep) allows the agent to recall previous inquiries, enabling it to provide more informed and contextually relevant responses over time.</p>



<p class="wp-block-paragraph"><strong>Use Case</strong>:</p>



<ul class="wp-block-list">
<li>Debugging code iteratively, where agents remember previous errors and solutions</li>
</ul>



<h2 class="wp-block-heading">Real-World Applications: AI Memory in Action</h2>



<p class="wp-block-paragraph">AI Agent Builders, equipped with memory systems, are transforming multiple industries by functioning much like human cognition:</p>



<ul class="wp-block-list">
<li><strong>Healthcare</strong> – AI agents remember patient history and medical records, much like doctors recalling past diagnoses.</li>



<li><strong>Education</strong> – AI-powered tutors personalize lessons by tracking student progress, similar to a teacher recognizing a student’s strengths and weaknesses.</li>



<li><strong>Customer Support</strong> – AI remembers previous conversations, ensuring continuity and a personalized experience, just as a helpful human assistant would.</li>



<li><strong>E-commerce</strong> – AI recalls user preferences and past purchases to offer tailored recommendations, much like a store clerk remembering a returning customer’s choices.</li>
</ul>



<h2 class="wp-block-heading">Conclusion: The Future of AI Memory</h2>



<p class="wp-block-paragraph">Just as human memory enables intelligence and decision-making, AI Agent Builders rely on memory to create smarter, more adaptive, and personalized systems. By learning from the human brain, AI memory architectures continue to evolve, pushing the boundaries of automation, personalization, and user experience.</p>



<p class="wp-block-paragraph">As AI memory systems improve, the line between human-like intelligence and artificial intelligence continues to blur. The key challenge? Ensuring these AI memories remain useful, ethical, and privacy-conscious—just like our own minds.</p>



<p class="wp-block-paragraph"><strong>References </strong>:</p>



<ol class="wp-block-list">
<li><a href="https://docs.crewai.com/concepts/memory" data-type="link" data-id="https://docs.crewai.com/concepts/memory" rel="nofollow noopener" target="_blank">Crew AI Documentation </a></li>



<li><a href="https://microsoft.github.io/autogen/0.2/docs/notebooks/agent_memory_using_zep" data-type="link" data-id="https://microsoft.github.io/autogen/0.2/docs/notebooks/agent_memory_using_zep" rel="nofollow noopener" target="_blank">Autogen Notebooks</a></li>
</ol>
]]></content:encoded>
					
					<wfw:commentRss>https://rpabotsworld.com/memory-in-ai-agents/feed/</wfw:commentRss>
			<slash:comments>1</slash:comments>
		
		
		<media:thumbnail url="https://rpabotsworld.com/wp-content/uploads/2025/02/Entities-in-the-Data-Service.jpg" />	</item>
		<item>
		<title>A Beginner’s Guide to Agentic AI: Working with CrewAI</title>
		<link>https://rpabotsworld.com/working-with-crewai/</link>
					<comments>https://rpabotsworld.com/working-with-crewai/#comments</comments>
		
		<dc:creator><![CDATA[Satish Prasad]]></dc:creator>
		<pubDate>Tue, 11 Feb 2025 06:08:35 +0000</pubDate>
				<category><![CDATA[Agentic AI & AI Automation]]></category>
		<category><![CDATA[crewai]]></category>
		<guid isPermaLink="false">https://rpabotsworld.com/?p=30163</guid>

					<description><![CDATA[Hey there, tech enthusiasts! 👋 I’m super excited to kick off my&#160;50-day learning journey into Agentic AI, and I’m starting today with&#160;CrewAI—a powerful open-source Python framework designed to orchestrate role-playing, autonomous AI agents that collaborate to tackle complex tasks. 🤖 If you’re curious about&#160;Agentic Automation&#160;and want to make a career shift from&#160;RPA to Agentic AI, [&#8230;]]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Hey there, tech enthusiasts! <img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f44b.png" alt="👋" class="wp-smiley" style="height: 1em; max-height: 1em;" /></p>



<p class="wp-block-paragraph">I’m super excited to kick off my&nbsp;<strong>50-day learning journey into Agentic AI</strong>, and I’m starting today with&nbsp;<strong>CrewAI</strong>—a powerful open-source Python framework designed to orchestrate role-playing, autonomous AI agents that collaborate to tackle complex tasks. <img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f916.png" alt="🤖" class="wp-smiley" style="height: 1em; max-height: 1em;" /></p>



<p class="wp-block-paragraph">If you’re curious about&nbsp;<strong>Agentic Automation</strong>&nbsp;and want to make a career shift from&nbsp;<strong>RPA to Agentic AI</strong>, this is the perfect time to join me! Let’s dive into the world of AI-driven collaboration and automation together. <img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f31f.png" alt="🌟" class="wp-smiley" style="height: 1em; max-height: 1em;" /></p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h2 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/2728.png" alt="✨" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <strong>What is CrewAI?</strong></h2>



<p class="wp-block-paragraph">CrewAI is an open-source framework that allows you to create and manage&nbsp;<strong>autonomous AI agents</strong>&nbsp;that work together to solve complex problems. By assigning specific roles, goals, and tools to each agent, CrewAI enables seamless collaboration, making it a game-changer for automation and AI-driven workflows.</p>



<p class="wp-block-paragraph">Here’s what makes CrewAI stand out:</p>



<figure class="wp-block-image size-full"><img decoding="async" width="2128" height="1460" src="https://rpabotsworld.com/wp-content/uploads/2025/02/crewAI-mindmap.png" alt="A Beginner’s Guide to Agentic AI: Working with CrewAI 3" class="wp-image-30166" title="https://github.com/crewAIInc/crewAI/blob/main/docs/crewAI-mindmap.png"></figure>



<h3 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f449.png" alt="👉" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <strong>Key Features of CrewAI</strong></h3>



<ul class="wp-block-list">
<li><strong>Role-Based Architecture</strong>: Each agent is assigned a distinct role and goal, ensuring specialized task execution.</li>



<li><strong>Agent Orchestration</strong>: Multiple agents work cohesively towards common objectives.</li>



<li><strong>Tool Integration</strong>: Agents can use external tools to extend their capabilities.</li>



<li><strong>Scalability</strong>: Whether you’re working on a small project or a complex system, CrewAI adapts to your needs.</li>
</ul>



<ul class="wp-block-list">
<li><strong>Production-Grade Architecture:</strong> Support for both high-level abstractions and low-level customization, with robust error handling and state management.</li>



<li><strong>Model Flexibility:</strong> Integrate with both OpenAI and open-source models, providing versatility in deployment.</li>



<li><strong>Event-Driven Flows:</strong> Build complex, real-world workflows with precise control over execution paths, state management, and conditional logic.</li>
</ul>



<p class="wp-block-paragraph">CrewAI&#8217;s modular architecture makes it ideal for scaling AI systems across various industries. Agents, tasks, and tools are reusable components that can be combined in various configurations to solve complex problems. Whether you&#8217;re automating workflows in finance, orchestrating research in academia, or managing customer queries in e-commerce, CrewAI provides a consistent, extensible framework.</p>



<h2 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4aa.png" alt="💪" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <strong>Getting Started with CrewAI</strong></h2>



<h3 class="wp-block-heading"><strong>Installation</strong></h3>



<p class="wp-block-paragraph">To get started, ensure you have&nbsp;<strong>Python (version 3.10 to 3.13)</strong>&nbsp;installed. Then, install CrewAI using pip:</p>



<pre class="wp-block-code"><code>pip install crewai</code></pre>



<p class="wp-block-paragraph">For additional tools, install with:</p>



<pre class="wp-block-code"><code>pip install 'crewai&#91;tools]'</code></pre>



<h2 class="wp-block-heading"><strong>Understanding Agents in CrewAI</strong></h2>



<p class="wp-block-paragraph">An&nbsp;<strong>agent</strong>&nbsp;in CrewAI is a fundamental component designed to perform specific tasks within a multi-agent system. Each agent has:</p>



<ol start="1" class="wp-block-list">
<li><strong>Role</strong>: Defines the agent’s function (e.g., “Data Researcher”).</li>



<li><strong>Goal</strong>: Specifies the objective the agent aims to achieve.</li>



<li><strong>Backstory</strong>: Provides context to influence the agent’s behavior.</li>



<li><strong>Tools</strong>: Extends the agent’s capabilities (e.g., web search tools).</li>



<li><strong>Configuration</strong>: Agents can be configured using YAML files for modular setups.</li>
</ol>



<p class="wp-block-paragraph"></p>



<h2 class="wp-block-heading"><strong>Example: Defining an Agent</strong></h2>



<h5 class="wp-block-heading"><strong>Content Writer Agent</strong></h5>



<p class="wp-block-paragraph">This agent is responsible for generating high-quality content based on a given topic.</p>



<pre class="wp-block-code"><code>from crewai import Agent

content_writer = Agent(
    role="Content Writer",
    goal="Write engaging and informative blog posts on AI and technology trends.",
    backstory="A creative writer with a passion for technology and AI, known for crafting compelling narratives.",
    tools=&#91;SerperDevTool()],  # Tool for web research
    verbose=True
)</code></pre>



<h5 class="wp-block-heading"><strong>Content Editor Agent</strong></h5>



<p class="wp-block-paragraph">This agent reviews and refines the content generated by the Content Writer.</p>



<pre class="wp-block-code"><code>content_editor = Agent(
    role="Content Editor",
    goal="Review and polish blog posts to ensure clarity, accuracy, and engagement.",
    backstory="A meticulous editor with a sharp eye for detail and a knack for improving readability.",
    verbose=True
)</code></pre>



<p class="wp-block-paragraph"><strong><em>Agents can also be defined in a YAML configuration file (agents.yaml), which allows for a clear and organized structure:</em></strong></p>



<h2 class="wp-block-heading"><strong>Understanding Task in CrewAI</strong></h2>



<p class="wp-block-paragraph">In CrewAI, the&nbsp;<strong>Task class</strong>&nbsp;is the backbone of defining and managing units of work that agents execute. Each task is designed to be modular, flexible, and highly configurable, ensuring that agents can perform their duties efficiently and effectively. Let’s break down the&nbsp;<strong>key attributes</strong>&nbsp;of the Task class and how they work together to create powerful workflows.</p>



<p class="wp-block-paragraph">In CrewAI, a <strong>Task</strong> represents a unit of work executed by an agent, defined by key attributes:</p>



<ul class="wp-block-list">
<li><strong>Description</strong>: Explains the task’s purpose.</li>



<li><strong>Expected Output</strong>: Defines the desired result.</li>



<li><strong>Agent</strong>: Assigned to complete the task.</li>



<li><strong>Tools</strong>: Optional resources for execution.</li>



<li><strong>Async Execution</strong>: Specifies if the task runs asynchronously.</li>



<li><strong>Context</strong>: Provides relevant data from other tasks.</li>
</ul>



<h2 class="wp-block-heading"><strong>Example: Defining an Task</strong></h2>



<h5 class="wp-block-heading"><strong>Writing Task</strong></h5>



<p class="wp-block-paragraph">This task instructs the Content Writer to create a blog post on a specific topic.</p>



<pre class="wp-block-code"><code>from crewai import Task

write_blog_task = Task(
    description="Write a 1000-word blog post on the impact of Agentic AI on the future of automation.",
    expected_output="A well-structured, engaging, and informative blog post in markdown format.",
    agent=content_writer,
    tools=&#91;search_tool],  # Tool for research
    async_execution=False
)</code></pre>



<h5 class="wp-block-heading"><strong>Editing Task</strong></h5>



<p class="wp-block-paragraph">This task instructs the Content Editor to review and refine the blog post.</p>



<pre class="wp-block-code"><code>edit_blog_task = Task(
    description="Review and edit the blog post written by the Content Writer to ensure it meets quality standards.",
    expected_output="A polished and error-free blog post ready for publication.",
    agent=content_editor,
    async_execution=False
)</code></pre>



<p class="wp-block-paragraph"><strong><em>Tasks can also be defined in a YAML configuration file (tasks.yaml), allowing for organized and modular setups:</em></strong></p>



<h2 class="wp-block-heading"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f512.png" alt="🔒" class="wp-smiley" style="height: 1em; max-height: 1em;" /> <strong>LLM Class in CrewAI</strong></h2>



<p class="wp-block-paragraph">The&nbsp;<strong>LLM class</strong>&nbsp;is pivotal for configuring the language models that agents use to process and generate text. It provides flexibility in connecting to various LLM providers, including&nbsp;<strong>OpenAI-compatible models</strong>&nbsp;and&nbsp;<strong>local models via Ollama</strong>.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h3 class="wp-block-heading"><strong>Key Features of the LLM Class</strong></h3>



<h4 class="wp-block-heading"><strong>1. Model Specification</strong></h4>



<p class="wp-block-paragraph">You can define the model name, such as:</p>



<ul class="wp-block-list">
<li><code>"gpt-4"</code></li>



<li><code>"gpt-3.5-turbo"</code></li>



<li><code>"ollama/llama3.1"</code></li>
</ul>



<h4 class="wp-block-heading"><strong>2. Provider Configuration</strong></h4>



<p class="wp-block-paragraph">Set up connections to different LLM providers by specifying:</p>



<ul class="wp-block-list">
<li><strong>API Key</strong>: For authenticated access.</li>



<li><strong>Base URL</strong>: For custom endpoints (e.g., OpenAI or local Ollama servers).</li>
</ul>



<h4 class="wp-block-heading"><strong>3. Parameter Customization</strong></h4>



<p class="wp-block-paragraph">Fine-tune the model’s behavior using parameters like:</p>



<ul class="wp-block-list">
<li><strong>Temperature</strong>: Controls the randomness of the output.</li>



<li><strong>Max Tokens</strong>: Limits the length of the generated response.</li>



<li><strong>Top_p</strong>: Applies nucleus sampling to limit the token selection pool.</li>
</ul>



<p class="wp-block-paragraph"><strong>Example: Configuring an LLM</strong><br></p>



<pre class="wp-block-code"><code>from crewai import LLM

openai_llm = LLM(
    model="gpt-4",
    api_key="your-openai-api-key",
    api_base="https://api.openai.com/v1"
)</code></pre>



<h3 class="wp-block-heading"><strong>Integrating LLMs with Agents</strong></h3>



<p class="wp-block-paragraph">Agents in CrewAI can be assigned specific LLMs to perform their tasks. For example:</p>



<pre class="wp-block-code"><code>from crewai import Agent

researcher = Agent(
    role="Researcher",
    goal="Conduct in-depth research on AI developments.",
    llm=openai_llm  # Assigning the OpenAI LLM
)</code></pre>



<h3 class="wp-block-heading"><strong>Configuring LLMs via Environment Variables</strong></h3>



<p class="wp-block-paragraph">Alternatively, you can configure LLMs using environment variables for dynamic and secure setups:</p>



<pre class="wp-block-code"><code>import os

os.environ&#91;"OPENAI_API_KEY"] = "your-openai-api-key"
os.environ&#91;"OPENAI_API_BASE"] = "https://api.openai.com/v1"</code></pre>



<h2 class="wp-block-heading"><strong>Crew Class in CrewAI</strong></h2>



<p class="wp-block-paragraph">The&nbsp;<strong>Crew class</strong>&nbsp;is the central component that orchestrates the collaboration of multiple agents and tasks. It manages the sequence and interaction of tasks, ensuring efficient and organized workflows.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h3 class="wp-block-heading"><strong>Key Components of the Crew Class</strong></h3>



<h4 class="wp-block-heading"><strong>1. Agents</strong></h4>



<p class="wp-block-paragraph">A list of&nbsp;<strong>Agent instances</strong>&nbsp;that define the roles and responsibilities within the crew.</p>



<h4 class="wp-block-heading"><strong>2. Tasks</strong></h4>



<p class="wp-block-paragraph">A list of&nbsp;<strong>Task instances</strong>&nbsp;that outline the specific actions to be executed.</p>



<h4 class="wp-block-heading"><strong>3. Process Flow</strong></h4>



<p class="wp-block-paragraph">Determines the execution strategy of tasks, such as:</p>



<ul class="wp-block-list">
<li><strong>Sequential</strong>: Tasks are executed one after another.</li>



<li><strong>Hierarchical</strong>: A manager agent coordinates the crew, delegating tasks and validating outcomes.</li>
</ul>



<h4 class="wp-block-heading"><strong>4. Verbose Mode</strong></h4>



<p class="wp-block-paragraph">Enables detailed logging of the crew’s operations for monitoring and debugging.</p>



<h2 class="wp-block-heading">Example <strong>: Orchestrating the Crew</strong><br></h2>



<p class="wp-block-paragraph">Now, let’s create a&nbsp;<strong>Crew</strong>&nbsp;to manage these agents and tasks.</p>



<pre class="wp-block-code"><code>from crewai import Crew, Process

content_creation_crew = Crew(
    agents=&#91;content_writer, content_editor],
    tasks=&#91;write_blog_task, edit_blog_task],
    process=Process.sequential,  # Tasks will be executed one after another
    verbose=True
)

# Execute the crew
final_output = content_creation_crew.kickoff()
print(final_output)</code></pre>



<p class="wp-block-paragraph">The&nbsp;kickoff()&nbsp;method triggers the execution of tasks as per the defined process flow.</p>



<p class="wp-block-paragraph"><strong>Additional Configurations:</strong></p>



<ul class="wp-block-list">
<li><strong>Output Logging</strong>: Specify an output log file to record the crew’s execution details.</li>



<li><strong>Token Usage Monitoring</strong>: Access the&nbsp;usage_metrics&nbsp;attribute post-execution to review language model usage metrics.</li>



<li><strong>Memory Utilization</strong>: Implement memory mechanisms to enhance the crew’s performance over time.</li>
</ul>



<p class="wp-block-paragraph">For comprehensive details on configuring and utilizing the&nbsp;Crew&nbsp;class, refer to the&nbsp;<a href="https://docs.crewai.com/concepts/crews" rel="nofollow noopener" target="_blank">CrewAI Documentation on Crews</a></p>



<h3 class="wp-block-heading"><strong>How It Works</strong></h3>



<ol start="1" class="wp-block-list">
<li>The&nbsp;<strong>Content Writer</strong>&nbsp;researches and writes the blog post.</li>



<li>The&nbsp;<strong>Content Editor</strong>&nbsp;reviews and polishes the blog post.</li>



<li>The&nbsp;<strong>Crew</strong>&nbsp;ensures the tasks are executed in sequence, producing a high-quality, publication-ready blog post.</li>
</ol>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<pre class="wp-block-code"><code>import os
from crewai import Agent, Task, Crew, Process
from crewai_tools import SerperDevTool,WebsiteSearchTool,DirectoryReadTool,FileReadTool

# Set up API keys
os.environ&#91;"SERPER_API_KEY"] = "" # serper.dev API key
os.environ&#91;"OPENAI_API_KEY"] = ""

# Instantiate tools
docs_tool = DirectoryReadTool(directory='./blog-posts')
file_tool = FileReadTool()
search_tool = SerperDevTool()
web_rag_tool = WebsiteSearchTool()


# Define your agents with roles and goals
content_writer = Agent(
    role="Content Writer",
    goal="Write engaging and informative blog posts on AI and technology trends.",
    backstory="A creative writer with a passion for technology and AI, known for crafting compelling narratives.",
    tools=&#91;search_tool,web_rag_tool],  # Tool for web research
    verbose=True
)

content_editor = Agent(
    role="Content Editor",
    goal="Review and polish blog posts to ensure clarity, accuracy, and engagement.",
    backstory="A meticulous editor with a sharp eye for detail and a knack for improving readability.",
    tools=&#91;docs_tool, file_tool],
    verbose=True
)


# Create tasks for your agents
write_blog_task = Task(
    description="Write a 1000-word blog post on the impact of Agentic AI on the future of automation.",
    expected_output="A well-structured, engaging, and informative blog post in markdown format.",
    agent=content_writer,
    tools=&#91;search_tool],  # Tool for research
    async_execution=False
)

edit_blog_task = Task(
    description="Review and edit the blog post written by the Content Writer to ensure it meets quality standards.",
    expected_output="A polished and error-free blog post ready for publication.",
    agent=content_editor,
    async_execution=False
)

# Instantiate your crew with a sequential process
content_creation_crew = Crew(
    agents=&#91;content_writer, content_editor],
    tasks=&#91;write_blog_task, edit_blog_task],
    process=Process.sequential,  # Tasks will be executed one after another
    verbose=True
)

# Execute the crew
final_output = content_creation_crew.kickoff()
print(final_output)</code></pre>



<h3 class="wp-block-heading"><strong>Why This Example?</strong></h3>



<ul class="wp-block-list">
<li><strong>Real-World Application</strong>: This workflow mimics real-world content creation processes.</li>



<li><strong>Role Specialization</strong>: Each agent has a clear role and goal, ensuring efficiency.</li>



<li><strong>Scalability</strong>: You can add more agents (e.g., a Graphic Designer for visuals) or tasks (e.g., SEO optimization) as needed.</li>
</ul>



<h2 class="wp-block-heading">! Important </h2>



<p class="wp-block-paragraph">The&nbsp;<strong>SerperDevTool</strong>&nbsp;is a powerful utility designed to perform semantic searches across the internet using the&nbsp;<strong>serper.dev API</strong>. It allows developers to fetch the most relevant search results based on a user-provided query, with options for customization such as localization, country-specific results, and more. Below is a comprehensive guide to help you get started with the tool, including installation, setup, and usage examples.</p>



<hr class="wp-block-separator has-alpha-channel-opacity"/>



<h3 class="wp-block-heading"><strong>Key Features</strong></h3>



<ol start="1" class="wp-block-list">
<li><strong>Semantic Search</strong>: Performs context-aware searches to return highly relevant results.</li>



<li><strong>Customization</strong>: Supports parameters like&nbsp;<code>country</code>,&nbsp;<code>location</code>,&nbsp;<code>locale</code>, and&nbsp;<code>n_results</code>&nbsp;for tailored searches.</li>



<li><strong>Easy Integration</strong>: Simple installation and setup process.</li>



<li><strong>Real-Time Results</strong>: Fetches up-to-date information from the internet.</li>
</ol>



<h3 class="wp-block-heading"><strong>Setup</strong></h3>



<ol start="1" class="wp-block-list">
<li><strong>Obtain an API Key</strong>:
<ul class="wp-block-list">
<li>Register for a free account at&nbsp;<a href="https://serper.dev/" target="_blank" rel="noreferrer noopener nofollow">serper.dev</a>&nbsp;to acquire your API key.</li>
</ul>
</li>



<li><strong>Configure Environment Variables</strong>:
<ul class="wp-block-list">
<li>Store your API key in an environment variable named&nbsp;<code>SERPER_API_KEY</code>.</li>
</ul>
</li>



<li></li>
</ol>



<h2 class="wp-block-heading"><strong>Join Me on This Journey!</strong></h2>



<p class="wp-block-paragraph">If you’re excited about&nbsp;<strong>Agentic AI</strong>&nbsp;and want to learn how to build such workflows, follow along with my 50-day journey! <img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f680.png" alt="🚀" class="wp-smiley" style="height: 1em; max-height: 1em;" /></p>



<ul class="wp-block-list">
<li><strong>Like</strong>&nbsp;this post if you found this breakdown helpful!</li>



<li><strong>Share</strong>&nbsp;it with your network to spread the knowledge.</li>



<li><strong>Follow</strong>&nbsp;me for daily updates and insights.</li>



<li><strong>Comment</strong>&nbsp;below if you’d like to see more examples like this!</li>
</ul>



<p class="wp-block-paragraph"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4cc.png" alt="📌" class="wp-smiley" style="height: 1em; max-height: 1em;" />&nbsp;<strong>Hashtags:</strong><br>#AgenticAI #CrewAI #LLM #AIWorkflows #TaskOrchestration #AIJourney #RPAToAgentic #CareerShift #LearnAI #TechTransformation</p>



<p class="wp-block-paragraph"><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4cc.png" alt="📌" class="wp-smiley" style="height: 1em; max-height: 1em;" />&nbsp;<strong>Additional Resources:</strong></p>



<p class="wp-block-paragraph">Let’s build the future of automation together! <img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f4aa.png" alt="💪" class="wp-smiley" style="height: 1em; max-height: 1em;" /><img src="https://s.w.org/images/core/emoji/17.0.2/72x72/1f525.png" alt="🔥" class="wp-smiley" style="height: 1em; max-height: 1em;" /></p>
]]></content:encoded>
					
					<wfw:commentRss>https://rpabotsworld.com/working-with-crewai/feed/</wfw:commentRss>
			<slash:comments>1</slash:comments>
		
		
		<media:thumbnail url="https://rpabotsworld.com/wp-content/uploads/2023/05/indian-software-developer-XAA6LDA.jpg" />	</item>
	</channel>
</rss>

<!--
Performance optimized by W3 Total Cache. Learn more: https://www.boldgrid.com/w3-total-cache/?utm_source=w3tc&utm_medium=footer_comment&utm_campaign=free_plugin

Page Caching using Disk: Enhanced 

Served from: rpabotsworld.com @ 2026-11-20 11:29:16 by W3 Total Cache
-->