<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Sulabh Sethi — AI/ML &amp; Defence Tech</title><description>Build logs, ML experiments, and applied AI for government and defence.</description><link>https://sulabhsethi.tech/</link><item><title>Build Log: Nyaya Mitra, Plain-Language Search for India&apos;s New Criminal Laws</title><link>https://sulabhsethi.tech/blog/build-log-nyaya-mitra/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/build-log-nyaya-mitra/</guid><description>A case study of something I actually shipped: a tool that lets an officer ask a question in plain words and get the right section of the new BNS, BNSS, and BSA, explained simply. Here is how it works and why I kept it boring.</description><pubDate>Wed, 17 Jun 2026 00:00:00 GMT</pubDate><category>Build Log</category><category>RAG</category><category>Legal Tech</category><category>Policing</category><category>Postgres</category></item><item><title>How PM GatiShakti Actually Works: The Architecture, Explained Simply</title><link>https://sulabhsethi.tech/blog/how-gatishakti-works/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/how-gatishakti-works/</guid><description>After the Bhashini posts, a reader asked for the same treatment of PM GatiShakti. It is a very different kind of platform: not AI models, but one shared map for planning India&apos;s infrastructure. A simple, fact-checked explainer.</description><pubDate>Mon, 15 Jun 2026 00:00:00 GMT</pubDate><category>GatiShakti</category><category>GIS</category><category>Government</category><category>Infrastructure</category><category>Explainer</category></item><item><title>AI Without a GPU: What a 16GB VPS Can Actually Run</title><link>https://sulabhsethi.tech/blog/ai-without-a-gpu-16gb-vps/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/ai-without-a-gpu-16gb-vps/</guid><description>You do not need an expensive GPU server to do real, useful AI. A practical, honest look at what a single modest machine with no GPU can run, what it cannot, and how I host a whole platform on one small box.</description><pubDate>Fri, 12 Jun 2026 00:00:00 GMT</pubDate><category>Self-Hosting</category><category>AI</category><category>Infrastructure</category><category>Practical Guide</category><category>Edge AI</category></item><item><title>How Bhashini Actually Works: The Architecture, Explained Simply</title><link>https://sulabhsethi.tech/blog/how-bhashini-works/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/how-bhashini-works/</guid><description>Bhashini lets you speak in one Indian language and be understood in another. People assume it is one giant AI. It is not. A simple, fact-checked look at the real models behind it, the base model for translation, and how they were trained.</description><pubDate>Wed, 10 Jun 2026 00:00:00 GMT</pubDate><category>Bhashini</category><category>AI4Bharat</category><category>Machine Translation</category><category>Indian Languages</category><category>Explainer</category></item><item><title>How RAG Works, and Why It Stops AI From Making Things Up</title><link>https://sulabhsethi.tech/blog/how-rag-works/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/how-rag-works/</guid><description>Large language models will confidently invent a law section that does not exist. Retrieval-Augmented Generation (RAG) is the fix. A plain-English explainer, grounded in a real legal-lookup tool I built for police.</description><pubDate>Tue, 09 Jun 2026 00:00:00 GMT</pubDate><category>RAG</category><category>LLM</category><category>AI</category><category>Legal Tech</category><category>Explainer</category></item><item><title>How an Investigating Officer Can Actually Use Machine Learning: A Practical Guide</title><link>https://sulabhsethi.tech/blog/machine-learning-for-investigating-officers/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/machine-learning-for-investigating-officers/</guid><description>Forget the hype. A grounded, plain-English guide to how the boring, explainable machine learning models can help a real investigating officer sort a pile of cases, spot the odd one out, and defend the reasoning, without ever pretending to be a detective.</description><pubDate>Mon, 08 Jun 2026 00:00:00 GMT</pubDate><category>Machine Learning</category><category>Policing</category><category>Practical Guide</category><category>Investigation</category><category>Explainer</category></item><item><title>Decision Trees, Random Forests, XGBoost: The Boring ML That Runs the World</title><link>https://sulabhsethi.tech/blog/decision-trees-random-forests-xgboost/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/decision-trees-random-forests-xgboost/</guid><description>Everyone talks about neural networks and LLMs. But the models quietly running your bank, your fraud alerts, and half of Kaggle are older, simpler, and far more boring. A plain-English tour of the real workhorses.</description><pubDate>Sun, 07 Jun 2026 00:00:00 GMT</pubDate><category>Machine Learning</category><category>Decision Trees</category><category>Random Forest</category><category>XGBoost</category><category>Explainer</category></item><item><title>AI vs Machine Learning vs Neural Networks vs LLMs: What&apos;s the Difference?</title><link>https://sulabhsethi.tech/blog/ai-ml-neural-networks-llms/</link><guid isPermaLink="true">https://sulabhsethi.tech/blog/ai-ml-neural-networks-llms/</guid><description>A plain-English guide to how AI, machine learning, neural networks, and large language models actually relate. They&apos;re not competing ideas — they&apos;re nested inside each other. Written by an ex-Army officer who now builds them.</description><pubDate>Sat, 06 Jun 2026 00:00:00 GMT</pubDate><category>AI</category><category>Machine Learning</category><category>Neural Networks</category><category>LLM</category><category>Explainer</category></item></channel></rss>