<?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 Montasir Mahmud on Medium]]></title>
        <description><![CDATA[Stories by Montasir Mahmud on Medium]]></description>
        <link>https://medium.com/@mmcse19?source=rss-bc8dfb6cd823------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*D0sy9ec2mILCTl3O</url>
            <title>Stories by Montasir Mahmud on Medium</title>
            <link>https://medium.com/@mmcse19?source=rss-bc8dfb6cd823------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 02:50:29 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mmcse19/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[[Day 45/100] Hybrid Search for RAG: BM25 Finds What Your Vectors Miss]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pub.towardsai.net/day-45-100-hybrid-search-for-rag-bm25-finds-what-your-vectors-miss-afd953519871?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1672/1*Rt1qh4G5XJr93Fu2s-y62g.png" width="1672"></a></p><p class="medium-feed-snippet">Why embeddings blur error codes and endpoint names, how BM25 and Reciprocal Rank Fusion fix it in about 60 lines of Python, and when to&#x2026;</p><p class="medium-feed-link"><a href="https://pub.towardsai.net/day-45-100-hybrid-search-for-rag-bm25-finds-what-your-vectors-miss-afd953519871?source=rss-bc8dfb6cd823------2">Continue reading on Towards AI »</a></p></div>]]></description>
            <link>https://pub.towardsai.net/day-45-100-hybrid-search-for-rag-bm25-finds-what-your-vectors-miss-afd953519871?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/afd953519871</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[rags]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 22:01:01 GMT</pubDate>
            <atom:updated>2026-11-14T22:01:01.432Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[[Day 44/100] RAG Chunking Strategies: Why Half Your Answer Lands in the Wrong Chunk]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pub.towardsai.net/day-44-100-rag-chunking-strategies-why-half-your-answer-lands-in-the-wrong-chunk-8fe985cb49d5?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1672/1*SYzeA4NpncbGlwrkHVbNuw.png" width="1672"></a></p><p class="medium-feed-snippet">Fixed, recursive, structure-aware, semantic, and hierarchical chunking, plus Anthropic&#x2019;s contextual retrieval, tested on one PTO policy&#x2026;</p><p class="medium-feed-link"><a href="https://pub.towardsai.net/day-44-100-rag-chunking-strategies-why-half-your-answer-lands-in-the-wrong-chunk-8fe985cb49d5?source=rss-bc8dfb6cd823------2">Continue reading on Towards AI »</a></p></div>]]></description>
            <link>https://pub.towardsai.net/day-44-100-rag-chunking-strategies-why-half-your-answer-lands-in-the-wrong-chunk-8fe985cb49d5?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/8fe985cb49d5</guid>
            <category><![CDATA[rags]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 21:01:01 GMT</pubDate>
            <atom:updated>2026-11-14T21:01:01.794Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[AI Could Pay Us All. History Says It Won’t Happen by Itself.]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@mmcse19/ai-could-pay-us-all-history-says-it-wont-happen-by-itself-92f515994740?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/2000/1*YznqSVvn1-6uFy9VG3SszA.jpeg" width="2000"></a></p><p class="medium-feed-snippet">In 1812, machines made Britain rich and left its workers waiting half a century for a share. AI is about to run the same experiment on the&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@mmcse19/ai-could-pay-us-all-history-says-it-wont-happen-by-itself-92f515994740?source=rss-bc8dfb6cd823------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@mmcse19/ai-could-pay-us-all-history-says-it-wont-happen-by-itself-92f515994740?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/92f515994740</guid>
            <category><![CDATA[superintelligence]]></category>
            <category><![CDATA[economics]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Fri, 13 Nov 2026 15:13:31 GMT</pubDate>
            <atom:updated>2026-11-13T15:13:31.460Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[[Day 43/100] Vector Database Comparison for RAG: pgvector, Qdrant, Pinecone, and More]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pub.towardsai.net/day-43-100-vector-database-comparison-for-rag-pgvector-qdrant-pinecone-and-more-b25ce34fb255?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1672/1*m66_ZGVXF9EkYDNeTdVbmA.png" width="1672"></a></p><p class="medium-feed-snippet">Six databases compared on where they run, how they filter, hybrid search, and tenant isolation, then a tested move from LangChain&#x2019;s&#x2026;</p><p class="medium-feed-link"><a href="https://pub.towardsai.net/day-43-100-vector-database-comparison-for-rag-pgvector-qdrant-pinecone-and-more-b25ce34fb255?source=rss-bc8dfb6cd823------2">Continue reading on Towards AI »</a></p></div>]]></description>
            <link>https://pub.towardsai.net/day-43-100-vector-database-comparison-for-rag-pgvector-qdrant-pinecone-and-more-b25ce34fb255?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/b25ce34fb255</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[rags]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Thu, 12 Nov 2026 21:01:01 GMT</pubDate>
            <atom:updated>2026-11-12T21:01:01.693Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[[Day 42/100] How to Choose an Embedding Model for RAG (Without Trusting the Leaderboard)]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pub.towardsai.net/day-42-100-how-to-choose-an-embedding-model-for-rag-without-trusting-the-leaderboard-1798657718c1?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1672/1*eKXSWL7qo9EqL2R3zs4HmQ.png" width="1672"></a></p><p class="medium-feed-snippet">Dimensions, Matryoshka truncation, token limits, query prefixes, and cost, plus a small recall@k harness that tests any LangChain&#x2026;</p><p class="medium-feed-link"><a href="https://pub.towardsai.net/day-42-100-how-to-choose-an-embedding-model-for-rag-without-trusting-the-leaderboard-1798657718c1?source=rss-bc8dfb6cd823------2">Continue reading on Towards AI »</a></p></div>]]></description>
            <link>https://pub.towardsai.net/day-42-100-how-to-choose-an-embedding-model-for-rag-without-trusting-the-leaderboard-1798657718c1?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/1798657718c1</guid>
            <category><![CDATA[rags]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Mon, 09 Nov 2026 14:01:02 GMT</pubDate>
            <atom:updated>2026-11-09T14:01:02.690Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Weight-Space Programming: One LLM Can Directly Edit Another LLM’s Weights]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/data-science-collective/weight-space-programming-how-editing-0-016-of-weights-boosted-math-scores-by-509-fe051c2a218a?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1672/1*xt-APbD6a6j1bnbInQI5uw.png" width="1672"></a></p><p class="medium-feed-snippet">GPT-6 Astra directly read and modified Qwen 0.8B&#x2019;s weights, changing just 0.016% of them and improving its math scores by 509%.</p><p class="medium-feed-link"><a href="https://medium.com/data-science-collective/weight-space-programming-how-editing-0-016-of-weights-boosted-math-scores-by-509-fe051c2a218a?source=rss-bc8dfb6cd823------2">Continue reading on Data Science Collective »</a></p></div>]]></description>
            <link>https://medium.com/data-science-collective/weight-space-programming-how-editing-0-016-of-weights-boosted-math-scores-by-509-fe051c2a218a?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/fe051c2a218a</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai-research]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Sun, 08 Nov 2026 17:44:37 GMT</pubDate>
            <atom:updated>2026-11-12T08:50:34.834Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[[Day 41/100] RAG Fundamentals: Why Your LLM Needs Retrieval, Not a Bigger Context Window]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pub.towardsai.net/day-41-100-rag-fundamentals-why-your-llm-needs-retrieval-not-a-bigger-context-window-d34826a68e2f?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1672/1*M8LTXBPv5iLGPt-ai6yXjw.png" width="1672"></a></p><p class="medium-feed-snippet">Long context and fine-tuning both look like shortcuts.</p><p class="medium-feed-link"><a href="https://pub.towardsai.net/day-41-100-rag-fundamentals-why-your-llm-needs-retrieval-not-a-bigger-context-window-d34826a68e2f?source=rss-bc8dfb6cd823------2">Continue reading on Towards AI »</a></p></div>]]></description>
            <link>https://pub.towardsai.net/day-41-100-rag-fundamentals-why-your-llm-needs-retrieval-not-a-bigger-context-window-d34826a68e2f?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/d34826a68e2f</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[rags]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Sat, 07 Nov 2026 16:01:02 GMT</pubDate>
            <atom:updated>2026-11-07T16:01:02.377Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[[Day 40/100] AI Agent Memory Privacy: Isolate Users, Redact PII, Delete for Real]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pub.towardsai.net/day-40-100-ai-agent-memory-privacy-isolate-users-redact-pii-delete-for-real-306329bb8a14?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1672/1*MpG0R3Olb2RsIWxjsSecqA.png" width="1672"></a></p><p class="medium-feed-snippet">One tool argument can leak every user&#x2019;s memories. How to scope identity, enforce row-level security in Postgres, keep PII out of storage&#x2026;</p><p class="medium-feed-link"><a href="https://pub.towardsai.net/day-40-100-ai-agent-memory-privacy-isolate-users-redact-pii-delete-for-real-306329bb8a14?source=rss-bc8dfb6cd823------2">Continue reading on Towards AI »</a></p></div>]]></description>
            <link>https://pub.towardsai.net/day-40-100-ai-agent-memory-privacy-isolate-users-redact-pii-delete-for-real-306329bb8a14?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/306329bb8a14</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[agentic-memory]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Wed, 04 Nov 2026 22:01:01 GMT</pubDate>
            <atom:updated>2026-11-06T12:44:31.459Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[AI Found a Few Old Bugs in PHP, and Europe Started Switching Off Its Own Websites]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@mmcse19/before-ai-makes-us-richer-it-might-make-the-internet-lock-its-doors-11d76a73bdef?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1774/1*T4HBa1z4dTBXelQMexuoYA.png" width="1774"></a></p><p class="medium-feed-snippet">In October, government websites across Europe went dark on purpose. The reason was a few old PHP bugs that AI helped dig up, and it says&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@mmcse19/before-ai-makes-us-richer-it-might-make-the-internet-lock-its-doors-11d76a73bdef?source=rss-bc8dfb6cd823------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@mmcse19/before-ai-makes-us-richer-it-might-make-the-internet-lock-its-doors-11d76a73bdef?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/11d76a73bdef</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[security]]></category>
            <category><![CDATA[internet]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Wed, 04 Nov 2026 04:59:15 GMT</pubDate>
            <atom:updated>2026-11-08T16:59:18.408Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[China Was Right About AI. It Still Lost $480 Billion.]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@mmcse19/china-was-right-about-ai-it-still-lost-480-billion-b910051b9742?source=rss-bc8dfb6cd823------2"><img src="https://cdn-images-1.medium.com/max/1619/1*enWrEGPE6dd7ZWKJ3q8NuA.png" width="1619"></a></p><p class="medium-feed-snippet">In under four months, China&#x2019;s hottest chip and robotics stocks gave back nearly half a trillion dollars. The robots are still walking. The&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@mmcse19/china-was-right-about-ai-it-still-lost-480-billion-b910051b9742?source=rss-bc8dfb6cd823------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@mmcse19/china-was-right-about-ai-it-still-lost-480-billion-b910051b9742?source=rss-bc8dfb6cd823------2</link>
            <guid isPermaLink="false">https://medium.com/p/b910051b9742</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[ai-bubble]]></category>
            <category><![CDATA[stock-market]]></category>
            <dc:creator><![CDATA[Montasir Mahmud]]></dc:creator>
            <pubDate>Tue, 03 Nov 2026 04:04:29 GMT</pubDate>
            <atom:updated>2026-11-03T04:04:29.898Z</atom:updated>
        </item>
    </channel>
</rss>