Vishal Kumar·Oct 15AI Governance — A Policy Is Not Governance Until Something Enforces ItWhat AI governance actually is, how it differs from the “Govern” pillar of an agent platform, and what it looks like when each sentence of…
Vishal Kumar·Oct 8Harness Engineering: LLM GuardrailsWhat a guardrail is, what LLM Guard and Microsoft Presidio give you, how to tune and test them, and why safety checks say nothing about a…
Vishal Kumar·Oct 5Can You Reward an LLM Every Time It Answers Correctly?What RLHF actually does, why “reward” only exists during training, and what to do with a thumbs-up button in the meantime.A response icon2A response icon2
Vishal Kumar·Oct 2The Answer Is in the Screenshot. Why Can’t Your RAG Find It? (Part 2: Finding and Showing Images)Part 1 got the words inside screenshots into the index. This part covers retrieving them, storing the bytes, showing the right image in an…
Vishal Kumar·Oct 1The Answer Is in the Screenshot. Why Can’t Your RAG Find It? (Part 1: Making Images Searchable)How to handle images in the wiki pages and runbooks people actually write. Part 1 covers getting image content into the index: parsing…
Vishal Kumar·Sep 30RAG — Confidence vs Cross-encoder Score vs GroundednessWhat a RAG “confidence” number actually measures, how it differs from the cross-encoder score, and why the useful one was already computed…
Vishal Kumar·Sep 29Why did my RAG system answer a Kubernetes question with a Docker runbook?A debugging story about bi-encoders, cross-encoders, HNSW, and the score my pipeline threw away.A response icon2A response icon2
Vishal Kumar·Aug 29SMTP: The Server Said “Sent.” The Inbox DisagreedA contact form on my site, PreReqs, said “Thanks — your enquiry is on its way.” every single time. The server answered every request with a…