LLMs don't think in words. I break down what happens between your prompt and the next token: tokenization, embeddings, attention, logits, and sampling. No math degree required, and no hype.
Hello, there 👋
Welcome to my blog! Built using NextJS, Tailwind CSS, ShadCN UI and Markdown.
Latest Posts
- Learn how to build an MCP (Model Context Protocol) server from scratch with zero dependencies. This step-by-step tutorial covers JSON-RPC, stdio transport, capability handshake, tool registration, and connecting to Claude Desktop and Cursor. Perfect for developers wanting to understand MCP internals without frameworks.
- Your model working in a notebook means nothing. I break down what it actually takes to deploy ML systems in production — the messy parts, the real tradeoffs, and what I wish I knew earlier.
- A practical 30-day roadmap to understand Machine Learning from first principles — covering data, models, evaluation, and real-world workflows without unnecessary math or hype.
- JWT, OAuth, and sessions aren’t competing — they solve different problems. I break down modern authentication the way it actually works in production, without buzzwords, hand-waving, or security theater.