Boris Cherny: production code written by Claude should clear a higher bar than human code, and he lists the guardrails Anthropic runs
Simon Willison's Weblog·medium signal
Simon Willison surfaced a 2026-09-11 quote from Boris Cherny arguing that Claude-written production code needs stricter oversight than human-written code, not looser. He names what Anthropic actually runs to hold that line: many lint rules, many tests, Claude-driven end-to-end tests, Claude-powered fuzzers running daily, automated code review, automated security review and automated refactoring. His stated consequence of skipping them is 'a mess that is hard to maintain down the line', which is a direct counter to the shipping-velocity framing.