Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: tokens
MCP Token Tax Re-Measured: Two Tool Descriptions Drove the Rise
Nine MCP servers re-measured 13 days later: tool definitions grew from 36,320 to 38,738 tokens, nearly all from one server rewriting its descriptions.
9 min readShrinking a Giant CLAUDE.md by 91% and Measuring What Broke
A 194K-character CLAUDE.md split into skills, references and path-scoped rules for Claude Code, checked for lost rules, then measured for how often skills fire.
8 min readRAG Without a Framework: Chunk, Score, Budget, Prompt
RAG explained as plain token selection: a short Python sketch with word-overlap scoring, a token budget gate, and what embeddings and vector stores really improve.
5 min readHow an LLM Works: Tokens, Attention and Next-Token Prediction
How an LLM works in plain terms: tokens, embeddings, self attention, causal masking and next-token generation, and why training and inference really differ.
8 min readWhere agentic coding tools spend tokens in practice
Context windows, tool loops, and retries multiply spend — budgets and caching that keep demos affordable.
9 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.