QA / Testing

Design, run, and score prompt evaluations with variance-aware benchmarks and regression tracking.

Generate Playwright E2E tests from user stories, with stable selectors and fixtures.

Audit and fix WCAG 2.2 AA issues: ARIA, colour contrast, keyboard nav, and screen reader support.

Drive a real browser to fill forms, scrape data, or test flows. Vision + DOM hybrid.

PR review in Claude. Style, security, perf, tests — with file-by-file inline comments.

Build, run, and report on LLM evals. Pairwise comparisons, judges, regression detection.

OWASP-aware code security audit — finds auth, input, secret, and dependency issues with remediation.

Run the app, drive the UI, confirm the change actually works — not just that tests pass.

Automate browser interactions — web testing, form filling, screenshots, scraping, multi-tab workflows.

Thorough PR review — security, performance, correctness, style, with inline-comment-ready output.

Deep multi-file code review powered by Fable 5's long-context reasoning — whole-PR understanding.