<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Prabhu K on Medium]]></title>
        <description><![CDATA[Stories by Prabhu K on Medium]]></description>
        <link>https://medium.com/@prabu-79668?source=rss-122b377ff72f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*GhIDhHTlSvi39QJcx7O8ew.png</url>
            <title>Stories by Prabhu K on Medium</title>
            <link>https://medium.com/@prabu-79668?source=rss-122b377ff72f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 08:20:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@prabu-79668/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Semantic Cache for LLMs and AI Agents: Cut Cost, Latency & Duplicate AI Work]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://prabu-79668.medium.com/semantic-cache-for-llms-and-ai-agents-cut-cost-latency-duplicate-ai-work-5e6185779ad8?source=rss-122b377ff72f------2"><img src="https://cdn-images-1.medium.com/max/1024/0*gHL4r9EM7UnsksnU.png" width="1024"></a></p><p class="medium-feed-snippet">Traditional caching asks, &#x201C;Have I seen this exact request before?&#x201D; Semantic caching asks a more useful question for generative AI: &#x201C;Have I&#x2026;</p><p class="medium-feed-link"><a href="https://prabu-79668.medium.com/semantic-cache-for-llms-and-ai-agents-cut-cost-latency-duplicate-ai-work-5e6185779ad8?source=rss-122b377ff72f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://prabu-79668.medium.com/semantic-cache-for-llms-and-ai-agents-cut-cost-latency-duplicate-ai-work-5e6185779ad8?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/5e6185779ad8</guid>
            <category><![CDATA[semantic-cache]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Wed, 02 Sep 2026 14:28:49 GMT</pubDate>
            <atom:updated>2026-09-02T14:28:49.803Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[What Is an Agentic AI Harness? Architecture, Components & Production Patterns]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://prabu-79668.medium.com/what-is-an-agentic-ai-harness-architecture-components-production-patterns-4234b78bab08?source=rss-122b377ff72f------2"><img src="https://cdn-images-1.medium.com/max/1024/0*lK_GL-Y8hhln0TO-.png" width="1024"></a></p><p class="medium-feed-snippet">From Powerful Models to Production-Ready Agents</p><p class="medium-feed-link"><a href="https://prabu-79668.medium.com/what-is-an-agentic-ai-harness-architecture-components-production-patterns-4234b78bab08?source=rss-122b377ff72f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://prabu-79668.medium.com/what-is-an-agentic-ai-harness-architecture-components-production-patterns-4234b78bab08?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/4234b78bab08</guid>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[agent-harness]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Sun, 09 Aug 2026 02:38:53 GMT</pubDate>
            <atom:updated>2026-08-09T02:38:53.196Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[15 Crucial AI Agent Design Patterns: A Practical Guide to Building Agentic Systems]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://prabu-79668.medium.com/15-crucial-ai-agent-design-patterns-a-practical-guide-to-building-agentic-systems-0064c7cca628?source=rss-122b377ff72f------2"><img src="https://cdn-images-1.medium.com/max/1536/0*8qG8F1e_blLomM1E.png" width="1536"></a></p><p class="medium-feed-snippet">AI agents are moving beyond simple prompt-and-response interactions. Modern agentic systems can plan, call tools, collaborate with&#x2026;</p><p class="medium-feed-link"><a href="https://prabu-79668.medium.com/15-crucial-ai-agent-design-patterns-a-practical-guide-to-building-agentic-systems-0064c7cca628?source=rss-122b377ff72f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://prabu-79668.medium.com/15-crucial-ai-agent-design-patterns-a-practical-guide-to-building-agentic-systems-0064c7cca628?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/0064c7cca628</guid>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[design-patterns]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Sun, 09 Aug 2026 02:21:25 GMT</pubDate>
            <atom:updated>2026-08-09T02:21:25.732Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Agentic AI — Design Patterns]]></title>
            <link>https://prabu-79668.medium.com/agentic-ai-design-patterns-9c6dca0164fd?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/9c6dca0164fd</guid>
            <category><![CDATA[ai-design-patterns]]></category>
            <category><![CDATA[design-patterns]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[agentic-ai-architecture]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Mon, 26 Jan 2026 20:57:02 GMT</pubDate>
            <atom:updated>2026-01-28T14:11:19.297Z</atom:updated>
            <content:encoded><![CDATA[<h3>Agentic AI — Design Patterns</h3><p>Large Language Models (LLMs) have evolved from passive text generators into <strong>active, goal-driven systems</strong> capable of reasoning, planning, and interacting with the world. This evolution has given rise to <strong>Agentic AI</strong> — AI systems that can observe, decide, act, and adapt autonomously.</p><p>As these systems mature, common architectural approaches have emerged. These <strong>design patterns</strong> help engineers build agentic systems that are modular, scalable, and robust.</p><p>This article outlines <strong>six foundational Agentic AI design patterns</strong>, each addressing a key capability required for intelligent, autonomous behavior.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g2evggzUVAdYD25CXwF8tA.png" /></figure><h3>Reflection Agent Pattern</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QunvAC57NzocD4f5qPCIWA.png" /></figure><p><strong>What it enables</strong><br> This pattern allows AI agents to pause, reflect, and evaluate their own behavior. By reasoning about past actions and outcomes, agents can adapt future strategies.</p><p><strong>Key capabilities</strong></p><ul><li>Self-monitoring and introspection</li><li>Evaluation of past decisions</li><li>Strategy refinement based on feedback</li></ul><p><strong>Why it matters</strong><br> Reflection improves reliability, reduces repeated errors, and supports long-horizon reasoning.</p><h4>Benefits of Reflection Pattern</h4><ul><li>Improved flexibility via self-examination</li><li>Applying historical performance knowledge</li><li>Error identification and correction</li><li>Improving the action and decision strategies</li><li>Supports adaption to manage challenges</li></ul><h3>Tool-Augmented (Tool-Using) Agent Pattern</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*elhvowBo1OAGzgSkNZKR7Q.png" /></figure><p><strong>What it enables</strong><br> LLMs overcome their static knowledge and execution limits by invoking external tools, APIs, and systems.</p><p><strong>Key capabilities</strong></p><ul><li>Access to real-time data</li><li>External computation and actions</li><li>Integration with enterprise systems</li></ul><p><strong>Why it matters</strong><br> This pattern transforms LLMs from “knowledge-only” models into <strong>capability-extended agents</strong>.</p><h4>Benefits of Tool-Augmented Pattern</h4><ul><li>Interaction with data sources and outside surroundings</li><li>Helps solve more varied and difficult problems</li><li>Increases effectiveness by using specialized tools</li><li>Promotes functional reusability and modularity</li></ul><h3>Planning Agent Pattern</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tnJsQD3nqZhs0X3nz7pPBw.png" /></figure><p><strong>What it enables</strong><br> Agents break complex objectives into smaller, manageable tasks and organize them into executable plans.</p><p><strong>Key capabilities</strong></p><ul><li>Goal decomposition</li><li>Step sequencing</li><li>Dependency-aware planning</li></ul><p><strong>Why it matters</strong><br> Without planning, agents struggle with complex, multi-step objectives</p><h4>Benefits of Planning Agent Pattern</h4><ul><li>Increases resilience with backup plans for blocks</li><li>Enables best action sequencing for efficiency</li><li>Task breakdown and goal-directed behavior</li><li>Improves the agent actions and reasoning</li><li>Helps multi-agent systems to coordinate</li></ul><h3>ReAct Agent Pattern</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*msNA1klM1BzeaURAZ2J0HQ.png" /></figure><p><strong>What it enables</strong><br> Agents dynamically alternate between reasoning and action, choosing what to think or do next based on observations.</p><p><strong>Key capabilities</strong></p><ul><li>Observation-driven decision-making</li><li>Interleaved reasoning and execution</li><li>Continuous adaptation to changing conditions</li></ul><p><strong>Why it matters</strong><br> Real-world environments are unpredictable. ReAct enables agents to remain flexible and responsive.</p><h4>Benefits of ReAct Agent Pattern</h4><ul><li><strong>Interleaves reasoning and action</strong>, allowing agents to make decisions based on real-time observations rather than static plans.</li><li><strong>Adapts naturally to dynamic environments</strong>, handling uncertainty and unexpected outcomes effectively.</li><li><strong>Reduces hallucination and over-planning</strong> by grounding reasoning in concrete actions and feedback.</li><li><strong>Excels at multi-step, tool-driven tasks</strong>, such as exploration, debugging, and interactive workflows.</li><li><strong>Improves transparency and debuggability</strong> through explicit, step-by-step reasoning–action traces</li></ul><h3>ReWOO Agent Pattern</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1sIX5j_UeJqB3gK0APIRyQ.png" /></figure><p><strong>What it enables</strong><br> This pattern decouples internal reasoning from external observations and tool calls using a modular architecture.</p><p><strong>Key capabilities</strong></p><ul><li>Separation of concerns</li><li>Clear reasoning vs execution boundaries</li><li>Easier debugging and extensibility</li></ul><p><strong>Why it matters</strong><br> Modularity improves maintainability and allows systems to evolve without entangling logic.</p><h4>Benefits of ReWOO Agent Pattern</h4><ul><li><strong>Separates reasoning from observation</strong>, allowing the agent to complete its reasoning phase without being influenced by intermediate tool outputs.</li><li><strong>Reduces reasoning errors and hallucinations</strong>, since plans are formed upfront rather than reactively rewritten after each observation.</li><li><strong>Improves efficiency and latency</strong>, as tool calls can be executed in parallel once the reasoning phase is finalized.</li><li><strong>Produces more stable and deterministic plans</strong>, making agent behavior easier to predict and test.</li><li><strong>Enhances debuggability and auditability</strong>, because the full reasoning trace exists independently of execution results</li></ul><h3>Multi-Agent Agent Pattern</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3uv4Sg9RJxTqGW80Pn3uNg.png" /></figure><p><strong>What it enables</strong><br> Multiple specialized agents collaborate to divide and execute complex tasks toward a shared objective.</p><p><strong>Key capabilities</strong></p><ul><li>Role specialization</li><li>Parallel execution</li><li>Coordination and communication</li></ul><p><strong>Why it matters</strong><br> Complex problems often exceed the capabilities of a single agent.</p><h4>Benefits of Multi-Agent Pattern</h4><ul><li><strong>Decomposes complex problems</strong> by distributing work across specialized agents, reducing cognitive and operational load on any single agent.</li><li><strong>Enables parallel execution</strong>, significantly improving throughput and time-to-completion for large or multi-step tasks.</li><li><strong>Supports specialization and role separation</strong>, allowing agents to focus on planning, reasoning, tool usage, or execution independently.</li><li><strong>Improves robustness and fault tolerance</strong>, as failures in one agent can be isolated or compensated for by others.</li><li><strong>Scales naturally with complexity</strong>, making the architecture suitable for enterprise-grade and long-running workflows.</li></ul><h3>Summary</h3><ul><li>A <strong>planning agent</strong> decomposes goals</li><li><strong>ReAct agents</strong> execute steps dynamically</li><li><strong>Tool-Planned agents</strong> interact with external systems</li><li><strong>Reflection agents</strong> improve performance over time</li><li><strong>ReWOO</strong> keep everything manageable</li><li><strong>Multi-agent collaboration</strong> scales complexity</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9c6dca0164fd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Rise of the Forward Deployment Engineer: The Hybrid Engineer Built for the AI Era]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-snippet">Technology roles evolve as fast as the technologies themselves&#x200A;&#x2014;&#x200A;but few roles have grown as rapidly or as quietly as the Forward&#x2026;</p><p class="medium-feed-link"><a href="https://prabu-79668.medium.com/the-rise-of-the-forward-deployment-engineer-the-hybrid-engineer-built-for-the-ai-era-7d794a537ac8?source=rss-122b377ff72f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://prabu-79668.medium.com/the-rise-of-the-forward-deployment-engineer-the-hybrid-engineer-built-for-the-ai-era-7d794a537ac8?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/7d794a537ac8</guid>
            <category><![CDATA[fde]]></category>
            <category><![CDATA[forward-deployed-engineer]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Mon, 01 Dec 2025 21:59:07 GMT</pubDate>
            <atom:updated>2025-12-01T21:59:07.716Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Microsoft Agent Framework vs AWS vs Google vs LangGraph: A Deep Technical Comparison of Enterprise…]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://prabu-79668.medium.com/microsoft-agent-framework-vs-aws-vs-google-vs-langgraph-a-deep-technical-comparison-of-enterprise-7d05d2e614b2?source=rss-122b377ff72f------2"><img src="https://cdn-images-1.medium.com/max/1536/1*0NqO6p32Ui1g5eKXc4AZoA.png" width="1536"></a></p><p class="medium-feed-snippet">A neutral, technical comparison of modern AI agent frameworks for enterprise leaders</p><p class="medium-feed-link"><a href="https://prabu-79668.medium.com/microsoft-agent-framework-vs-aws-vs-google-vs-langgraph-a-deep-technical-comparison-of-enterprise-7d05d2e614b2?source=rss-122b377ff72f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://prabu-79668.medium.com/microsoft-agent-framework-vs-aws-vs-google-vs-langgraph-a-deep-technical-comparison-of-enterprise-7d05d2e614b2?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/7d05d2e614b2</guid>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Wed, 12 Nov 2025 21:29:41 GMT</pubDate>
            <atom:updated>2025-11-12T21:29:41.807Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Build a Simple Agentic Flow with Open Source Components]]></title>
            <link>https://prabu-79668.medium.com/build-a-simple-agentic-flow-with-open-source-components-813691e73095?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/813691e73095</guid>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[langchain]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Wed, 23 Jul 2025 17:45:23 GMT</pubDate>
            <atom:updated>2025-07-23T17:45:23.912Z</atom:updated>
            <content:encoded><![CDATA[<p>Let’s create a <strong>weather assistant</strong> using agents and tools</p><p><strong>Step 1: Install Required Libraries</strong></p><pre>pip install langchain langgraph chromadb openai</pre><p>You can also install <strong>Ollama</strong> locally to use open LLMs like Mistral:</p><pre>brew install ollama<br>ollama run mistral</pre><p><strong>Step 2: Define a Tool</strong></p><pre>from dotenv import load_dotenv<br>import os<br><br># Load environment variables from a .env file<br>load_dotenv()<br>from langchain.agents import Tool<br><br>def get_weather(location: str) -&gt; str:<br>    return {<br>    &quot;location&quot;: location,<br>    &quot;result&quot;: f&quot;The weather in {location} is 30°C and sunny.&quot;<br>    }<br><br>tools = [<br>    Tool.from_function(<br>        func=get_weather,<br>        name=&quot;get_weather&quot;,<br>        description=&quot;Provides current weather information for a location.&quot;<br>    )<br>]</pre><p><strong>Step 3: Create a LangChain Agent</strong></p><pre>from langchain.chat_models import ChatOpenAI<br>from langchain.agents import initialize_agent, AgentType<br><br>llm = ChatOpenAI(temperature=0)<br><br>agent = initialize_agent(<br>    tools=tools,<br>    llm=llm,<br>    agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION,<br>    verbose=True<br>)<br><br>response = agent.run(&quot;What’s the weather in Bangalore?&quot;)<br>print(response)</pre><p><strong>Step 4: Orchestrate with LangGraph</strong></p><pre>from langgraph.graph import StateGraph, END<br>from typing import TypedDict<br><br># Step 1: Define your state<br>class WeatherState(TypedDict):<br>    location: str<br>    result: str<br>    <br>workflow = StateGraph(state_schema=WeatherState)<br><br>workflow.add_node(&quot;get_weather&quot;, lambda x: get_weather(x[&quot;location&quot;]))<br>workflow.set_entry_point(&quot;get_weather&quot;)<br>workflow.set_finish_point(&quot;get_weather&quot;)<br>workflow.add_edge(&quot;get_weather&quot;, END)  # correct use of END<br><br>graph = workflow.compile()<br>output = graph.invoke({&quot;location&quot;: &quot;Chennai&quot;})<br>print(output)<br><br>Note: Add a .env file and OPENAI_API_KEY from the openai.com</pre><p><strong>Sample Output</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Re9qkmA5-ZoeegqJPHhBrg.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=813691e73095" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Agentic Flow Using Open Source Components]]></title>
            <link>https://prabu-79668.medium.com/agentic-flow-using-open-source-components-cdaea1f7b01d?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/cdaea1f7b01d</guid>
            <category><![CDATA[agents]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Wed, 23 Jul 2025 15:41:26 GMT</pubDate>
            <atom:updated>2025-07-23T15:41:26.065Z</atom:updated>
            <content:encoded><![CDATA[<p>Agentic AI is reshaping how intelligent systems operate — enabling autonomous agents to collaborate, plan, and execute complex workflows without constant human intervention. In this article, we’ll explore how to <strong>build an agentic flow</strong> using <strong>open-source components</strong>, allowing you to create LLM-driven systems that are modular, explainable, and extensible.</p><p><strong>What Is an Agentic Flow?</strong></p><p>An <strong>agentic flow</strong> is a structured system where AI agents:</p><ul><li><strong>Plan</strong>: Break down a user goal into actionable tasks.</li><li><strong>Execute</strong>: Use tools, LLMs, or APIs to accomplish each task.</li><li><strong>Reason</strong>: Make decisions based on intermediate results.</li><li><strong>Loop</strong>: Iterate with memory and feedback to refine outcomes.</li></ul><p><strong>Key Components of an Agentic Flow</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R00cC1jWQnZwamyg1GM18w.png" /></figure><p><strong>Open Source Tech Stack</strong></p><p>Here’s a stack we’ll use to build our flow:</p><ul><li><strong>LLMs</strong>: Mistral 7B, Phi-3</li><li><strong>Runner</strong>: Ollama — run LLMs locally</li><li><strong>Agents &amp; Tools</strong>: LangChain</li><li><strong>Workflow Orchestration</strong>: <a href="https://github.com/langchain-ai/langgraph">LangGraph</a></li><li><strong>Memory</strong>: ChromaDB or FAISS</li><li><strong>Custom Tools</strong>: API clients, DB connectors, calculators, web searchers</li></ul><p>Reference</p><p><a href="https://docs.langchain.com/">https://docs.langchain.com/</a></p><p><a href="https://ollama.com/">https://ollama.com/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cdaea1f7b01d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unlocking Generative AI at Scale with Amazon Bedrock]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://prabu-79668.medium.com/unlocking-generative-ai-at-scale-with-amazon-bedrock-5dcda8acd982?source=rss-122b377ff72f------2"><img src="https://cdn-images-1.medium.com/max/2354/1*lB39LItQ6kmVCHLrJlf2iw.jpeg" width="2354"></a></p><p class="medium-feed-snippet">Introduction</p><p class="medium-feed-link"><a href="https://prabu-79668.medium.com/unlocking-generative-ai-at-scale-with-amazon-bedrock-5dcda8acd982?source=rss-122b377ff72f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://prabu-79668.medium.com/unlocking-generative-ai-at-scale-with-amazon-bedrock-5dcda8acd982?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/5dcda8acd982</guid>
            <category><![CDATA[amazon-bedrock]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Tue, 22 Jul 2025 20:09:39 GMT</pubDate>
            <atom:updated>2025-07-22T20:09:39.236Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Sample GenAI Platform: Step-by-Step]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://prabu-79668.medium.com/sample-genai-platform-step-by-step-eec12c1b3163?source=rss-122b377ff72f------2"><img src="https://cdn-images-1.medium.com/max/1024/1*ufWa2QDST5fS2jlTHRcdFA.png" width="1024"></a></p><p class="medium-feed-snippet">Here&#x2019;s a high-level workflow for building a chatbot platform using Azure OpenAI:</p><p class="medium-feed-link"><a href="https://prabu-79668.medium.com/sample-genai-platform-step-by-step-eec12c1b3163?source=rss-122b377ff72f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://prabu-79668.medium.com/sample-genai-platform-step-by-step-eec12c1b3163?source=rss-122b377ff72f------2</link>
            <guid isPermaLink="false">https://medium.com/p/eec12c1b3163</guid>
            <category><![CDATA[azureopenai]]></category>
            <dc:creator><![CDATA[Prabhu K]]></dc:creator>
            <pubDate>Mon, 21 Jul 2025 20:29:01 GMT</pubDate>
            <atom:updated>2025-07-21T20:29:01.836Z</atom:updated>
        </item>
    </channel>
</rss>