Playwright users–wants design-partner to shape an AI tool that tames flaky tests

Hi HN,

We’re 2 indie engineers building an AI-powered observability layer for Playwright. It clusters failures, spots flaky tests, and spits out plain-English root-cause hints so you can triage in minutes instead of combing through logs. Everything would be self-hosted—no data leaves your infra.

We’re aiming to open access mid-August and need 5–8 design partners to make sure we’re solving real pain.

You might be a fit if … • Your CI already runs Playwright (≈ 500+ tests or chronic flake pain) • You’re comfortable dropping in a lightweight agent • You can spare 2–3 short feedback calls over the next six weeks

What you’ll get • 6 months free on any plan • 50% lifetime discount on any plan in future • Direct influence on the roadmap (GitHub Checks, predictive alerts, etc.) • Fast turnaround on your feature requests—straight from the founders

Sound interesting? Comment below or ping the email in my HN profile and I’ll share a 2-minute Loom demo plus install instructions.

Thanks for reading!