Delivery and program leader with 15+ years on large-scale cloud programs, now building GenAI products and writing about AI — how it is built, how it is judged, and where it falls short. Currently a Manager at State Street — Charles River Development.https://kumarmukesh.pages.dev/en-ushttps://medium.com/@mukesh.itemail/p%C4%81%E1%B9%87inis-three-words-how-an-ancient-grammar-engine-baffled-the-modern-mind-a8a20033c465https://medium.com/@mukesh.itemail/p%C4%81%E1%B9%87inis-three-words-how-an-ancient-grammar-engine-baffled-the-modern-mind-a8a20033c465A 2,500-year-old Sanskrit grammar resolves rule conflicts with a single metarule stated in three words. Most system prompts I have read resolve theirs with a line in caps, added under pressure, that nobody has gone back to check. On what the older engine got right.Sun, 20 Sep 2026 00:00:00 GMTSystem PromptsAI ProductEssayhttps://medium.com/ai-engineering-simplified/evals-are-the-new-unit-tests-building-reliable-ai-products-6548e9cfe4eehttps://medium.com/ai-engineering-simplified/evals-are-the-new-unit-tests-building-reliable-ai-products-6548e9cfe4eeTraditional software tests assume one correct answer. AI systems do not work that way, which is why so many teams fall back on testing by hand and calling it quality. A practical case for building a repeatable evaluation process, and what it takes to run one with every important change.Thu, 20 Aug 2026 00:00:00 GMTEvaluationAI ProductQualityhttps://generativeai.pub/asset-assistant-building-an-ai-search-tool-that-finds-files-even-when-you-cannot-remember-their-5385349631fahttps://generativeai.pub/asset-assistant-building-an-ai-search-tool-that-finds-files-even-when-you-cannot-remember-their-5385349631faRegular search only looks at file names — it never looks at what is inside a file. A build write-up of an AI search tool that reads the file itself, why I deliberately avoided a generative model, and how I made sure every result points at something real.Fri, 14 Aug 2026 00:00:00 GMTAI SearchProduct BuildRetrieval