Benchmarking

2 articles

Web Engineering • Sep 25, 2026

Fix Performance Regressions With a Benchmark Ratchet (3.1x, 0 Breaks)

Fix performance regressions with a benchmark ratchet that fails CI on regression and locks in wins. How Claude shipped 3,000 changes at 3.1x, zero breaks.

9 min read

AI Coding Agents & DX • Aug 19, 2026

How to verify AI-generated benchmark claims: 1.4x to 1.5x slower

How to verify AI-generated benchmark claims: an LLM agent's regex engine beat Rust's regex crate by 1.4x on rebar, then lost 10x on a holdout it never saw.

9 min read