Bitwise Engineering

What you’ll find here

• λ

  Functional programming

  Types, purity and composition as everyday engineering tools: what they rule out, and what they cost.

• 01

  Succinct data structures

  Structures that answer queries in close to the minimum space the data needs: bit vectors, rank and select, and the engineering that makes them fast.

• AI

  AI workflows

  Working with coding agents and models in practice: what to hand them, how to check their work, and where the guardrails go.

Posts

• Explainer

  Announcing patchcov

  Patch coverage for git diffs, as a command and a GitHub Action: a precise, gateable answer to what a change left untested, with no service in the loop.

  draft2026-10-11
• Example · Decision

  ADR-002: Rate limit at the edge gateway

  Enforce one per-client token bucket in the gateway, with state in Redis, so limits hold however many services and replicas run behind it.

  accepted2026-10-05
• Reference

  Component gallery

  Every building block in the theme, with the markup to use it. The visual reference for authors and agents.

  accepted2026-10-05
• Reference

  Glossary

  Shared definitions for the terms used across these documents. Words defined here are underlined on every page; hover or tap one to see its definition.

  accepted2026-10-05
• Example · Design doc

  Moving rate limiting to the edge

  Replace per-service limiters with one token bucket per client in the gateway, backed by Redis, so a client's limit holds however far we scale.

  review2026-10-05
• Example · Explainer

  Token bucket rate limiting

  How a token bucket admits bursts but enforces an average rate — with a live simulation, implementations, and the mistakes that bite in production.

  accepted2026-10-05
• Example · Decision

  ADR-001: Rate limit inside each service

  Each service enforces its own per-client token bucket in process, so we can ship limits before launch without new infrastructure.

  superseded2025-11-12

  Replaced by ADR-002: Rate limit at the edge gateway