One engineer, team-scale output
Quality at scale cannot depend on how fast one person types. It has to come from a system. On Attorney Share, a single QA engineer built and runs a Playwright E2E suite — ~500 tests, ~25 areas — with Claude Code as the co-engineer.
The loop
Set the guardrails and let the AI drive: a written plan, small reviewable PRs, and verdicts that come from end-to-end runs against real preview environments. Reliability came with it — flaky tests fixed at the root, not masked with retries.
Read the full story in our case study.