<?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 Pranav Agrawal on Medium]]></title>
        <description><![CDATA[Stories by Pranav Agrawal on Medium]]></description>
        <link>https://medium.com/@pranav-agrawal?source=rss-8d4f8de5d103------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*JTbICM-h8lsrQqA-</url>
            <title>Stories by Pranav Agrawal on Medium</title>
            <link>https://medium.com/@pranav-agrawal?source=rss-8d4f8de5d103------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 00:30:09 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@pranav-agrawal/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[Learning AI Agents When AI Writes the Code: A CV Chatbot in Plain Python]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/learning-ai-agents-when-ai-writes-the-code-a-cv-chatbot-in-plain-python-2e29614f725b?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1440/1*Ohtsd9abk8rH8g6JmO2fcg.png" width="1440"></a></p><p class="medium-feed-snippet">When Claude and GPT write most of the code, how do you actually learn this stuff? I built a CV chatbot from scratch in plain Python, and&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/learning-ai-agents-when-ai-writes-the-code-a-cv-chatbot-in-plain-python-2e29614f725b?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/learning-ai-agents-when-ai-writes-the-code-a-cv-chatbot-in-plain-python-2e29614f725b?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/2e29614f725b</guid>
            <category><![CDATA[python-programming]]></category>
            <category><![CDATA[ollama]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[llm]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 11:42:07 GMT</pubDate>
            <atom:updated>2026-11-14T11:46:20.293Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Context Engineering Explained: What Goes Into an AI Agent’s Context Window]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/context-engineering-explained-what-goes-into-an-ai-agents-context-window-1388d1a427b1?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1344/1*qA3c_Vbnqt8KxviVN1R2pQ.png" width="1344"></a></p><p class="medium-feed-snippet">An AI agent can have the right tools, the right data and a well-written prompt, and still give a wrong answer forty turns into a task. The&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/context-engineering-explained-what-goes-into-an-ai-agents-context-window-1388d1a427b1?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/context-engineering-explained-what-goes-into-an-ai-agents-context-window-1388d1a427b1?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/1388d1a427b1</guid>
            <category><![CDATA[context-engineering]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[prompt-engineering]]></category>
            <category><![CDATA[large-language-models]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Mon, 26 Oct 2026 15:51:24 GMT</pubDate>
            <atom:updated>2026-10-26T15:51:24.330Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[MCP Explained: A Universal Standard for Connecting AI Agents to Tools]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://generativeai.pub/mcp-explained-a-universal-standard-for-connecting-ai-agents-to-tools-8fb10cd1e49a?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1677/1*yBMHu4Y64PIlLzqvtI9RWA.png" width="1677"></a></p><p class="medium-feed-snippet">How MCP servers, clients, and hosts work together, with practical examples and a simple Python implementation.</p><p class="medium-feed-link"><a href="https://generativeai.pub/mcp-explained-a-universal-standard-for-connecting-ai-agents-to-tools-8fb10cd1e49a?source=rss-8d4f8de5d103------2">Continue reading on Generative AI »</a></p></div>]]></description>
            <link>https://generativeai.pub/mcp-explained-a-universal-standard-for-connecting-ai-agents-to-tools-8fb10cd1e49a?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/8fb10cd1e49a</guid>
            <category><![CDATA[mcps]]></category>
            <category><![CDATA[model-context-protocol]]></category>
            <category><![CDATA[ai-engineering]]></category>
            <category><![CDATA[ai-agents-in-action]]></category>
            <category><![CDATA[llm]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Sat, 24 Oct 2026 12:03:07 GMT</pubDate>
            <atom:updated>2026-10-27T01:44:12.485Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The ReAct Pattern Explained: How AI Agents Reason Before They Act]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/the-react-pattern-explained-how-ai-agents-reason-before-they-act-8d91e84f776f?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1322/1*8vbOfpkKSRhB7l2-u4l2NQ.png" width="1322"></a></p><p class="medium-feed-snippet">In Part 3, function calling gave an agent the power to act. But being able to act is not the same as knowing what to do. ReAct closes that&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/the-react-pattern-explained-how-ai-agents-reason-before-they-act-8d91e84f776f?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/the-react-pattern-explained-how-ai-agents-reason-before-they-act-8d91e84f776f?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/8d91e84f776f</guid>
            <category><![CDATA[react]]></category>
            <category><![CDATA[agentic-coding]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[calling-function]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Fri, 23 Oct 2026 17:24:02 GMT</pubDate>
            <atom:updated>2026-10-23T17:24:02.855Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Function Calling Explained: How a Language Model Operates External Tools]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/function-calling-explained-how-a-language-model-operates-external-tools-601cde158483?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1568/1*9JFQ5ioktc4EhtNPAU5wvA.png" width="1568"></a></p><p class="medium-feed-snippet">A language model can only produce text. This article explains how that text becomes a reliable, structured instruction that your code can&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/function-calling-explained-how-a-language-model-operates-external-tools-601cde158483?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/function-calling-explained-how-a-language-model-operates-external-tools-601cde158483?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/601cde158483</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[calling-function]]></category>
            <category><![CDATA[ai-engineering]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[large-language-models]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Thu, 22 Oct 2026 17:27:10 GMT</pubDate>
            <atom:updated>2026-10-22T17:27:10.337Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Chatbot, Workflow, or Agent? A Decision Framework for LLM Applications]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/chatbot-workflow-or-agent-a-decision-framework-for-llm-applications-b5cbe8956b29?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1336/1*MgZ1Yvzyc1H0rEf5HBIHhA.png" width="1336"></a></p><p class="medium-feed-snippet">In Part 1 we found out an agent is just a model, some tools and a loop. Which raises an awkward follow-up: if it&#x2019;s that simple, why not&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/chatbot-workflow-or-agent-a-decision-framework-for-llm-applications-b5cbe8956b29?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/chatbot-workflow-or-agent-a-decision-framework-for-llm-applications-b5cbe8956b29?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/b5cbe8956b29</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai-engineering]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Mon, 19 Oct 2026 12:01:07 GMT</pubDate>
            <atom:updated>2026-10-19T12:01:07.488Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[What an AI Agent Actually Is: A Loop, a Model and Some Tools]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/what-an-ai-agent-actually-is-a-loop-a-model-and-some-tools-18df1fe04533?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/2556/1*57nQF5LCZW9vRmJYVNLr2Q.png" width="2556"></a></p><p class="medium-feed-snippet">&#x201C;AI agent&#x201D; is one of those phrases that sounds either like science fiction or like marketing, depending on who&#x2019;s saying it. But underneath&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/what-an-ai-agent-actually-is-a-loop-a-model-and-some-tools-18df1fe04533?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/what-an-ai-agent-actually-is-a-loop-a-model-and-some-tools-18df1fe04533?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/18df1fe04533</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[generative-ai-tools]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Sun, 18 Oct 2026 13:00:19 GMT</pubDate>
            <atom:updated>2026-10-18T13:00:19.761Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Gaussian Mixture Models: Guide to Soft Clustering]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/gaussian-mixture-models-guide-to-soft-clustering-2642422a3c11?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1332/1*aMnxcueM5QkB0VbMJZp5Qg.png" width="1332"></a></p><p class="medium-feed-snippet">Imagine you&#x2019;re handed a fish that could be a big salmon or a small tuna. Ask K-means and it will answer instantly, with total confidence&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/gaussian-mixture-models-guide-to-soft-clustering-2642422a3c11?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/gaussian-mixture-models-guide-to-soft-clustering-2642422a3c11?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/2642422a3c11</guid>
            <category><![CDATA[gaussian-mixture-model]]></category>
            <category><![CDATA[expectation-maximization]]></category>
            <category><![CDATA[kmeans-clustering]]></category>
            <category><![CDATA[clustering]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Sat, 18 Jul 2026 18:21:54 GMT</pubDate>
            <atom:updated>2026-07-18T18:23:23.219Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Classification Metrics: Why 99% Accuracy Can Still Be Useless]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/classification-metrics-why-99-accuracy-can-still-be-useless-970b706d606c?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/1330/1*LGUcBPJxMgM2ddIWmKRCbA.png" width="1330"></a></p><p class="medium-feed-snippet">A tour of accuracy, precision, recall, F1, and the two curves that finally tell you the truth for a classification model.</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/classification-metrics-why-99-accuracy-can-still-be-useless-970b706d606c?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/classification-metrics-why-99-accuracy-can-still-be-useless-970b706d606c?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/970b706d606c</guid>
            <category><![CDATA[recall]]></category>
            <category><![CDATA[precision]]></category>
            <category><![CDATA[f1]]></category>
            <category><![CDATA[pr-curve]]></category>
            <category><![CDATA[roc]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Thu, 11 Jun 2026 21:01:24 GMT</pubDate>
            <atom:updated>2026-06-17T16:00:17.631Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[How a Language Model Actually Reads: Tokenization from Scratch]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pranav-agrawal.medium.com/how-a-language-model-actually-reads-tokenization-from-scratch-189430b6eb24?source=rss-8d4f8de5d103------2"><img src="https://cdn-images-1.medium.com/max/2600/1*fy76a4eYdmbo-xIIFP1RwQ.png" width="3200"></a></p><p class="medium-feed-snippet">Transformers, Part 1. Before attention, before embeddings, before any of the parts everyone talks about, there&#x2019;s one unglamorous step that&#x2026;</p><p class="medium-feed-link"><a href="https://pranav-agrawal.medium.com/how-a-language-model-actually-reads-tokenization-from-scratch-189430b6eb24?source=rss-8d4f8de5d103------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://pranav-agrawal.medium.com/how-a-language-model-actually-reads-tokenization-from-scratch-189430b6eb24?source=rss-8d4f8de5d103------2</link>
            <guid isPermaLink="false">https://medium.com/p/189430b6eb24</guid>
            <category><![CDATA[tokenization]]></category>
            <category><![CDATA[text-to-token]]></category>
            <category><![CDATA[byte-pair-encoding]]></category>
            <category><![CDATA[transformers]]></category>
            <category><![CDATA[subword-tokenization]]></category>
            <dc:creator><![CDATA[Pranav Agrawal]]></dc:creator>
            <pubDate>Sat, 30 May 2026 17:26:02 GMT</pubDate>
            <atom:updated>2026-05-31T06:43:03.460Z</atom:updated>
        </item>
    </channel>
</rss>