Writing

Notes on building AI agents, the APIs they call, and making software people can trust. Also available as RSS.

TutorialSeptember 5, 2026ai-engineering

AI engineering starter kit: tools to learn first

The terminal, Linux, git, Python environments, Docker and VS Code setup you need before ML and AI work, explained plainly with commands that work.

12 min readRead

20265 posts

TutorialOct 24ragAdvanced RAG: hybrid search, reranking, agentsAdvanced RAG techniques that fix real retrieval failures: hybrid search with RRF, cross-encoder reranking, query rewriting, agentic retrieval and text-to-SQL.
TutorialOct 17ragRAG explained: build retrieval-augmented generationRAG explained for engineers: how to load, chunk, embed, store and retrieve documents, ground answers with citations, and evaluate and debug the pipeline.
EssayOct 10context-engineeringContext engineering: why the window beats the promptWhat context engineering is, how it differs from prompt engineering, and how to select, order and compress what goes in the window, plus a checklist.
TutorialOct 3prompt-engineeringPrompt engineering that holds up in productionPrompt engineering for production: roles, XML delimiters, few-shot, structured output, chaining, injection defence, and testing prompts with evals.
TutorialSep 19machine-learningMath for machine learning: what you actually needThe linear algebra, calculus and probability that machine learning uses, with short NumPy examples, what to skip, and a study order with free resources.