haf zhang

Finally a tool that treats flaky E2E tests as a code-gen problem, not a human debugging marathon! Love that the generated Playwright lives in my repo—no black-box lock-in and I can still npx playwright test on CI. Two quick questions: Does the AI auto-heal only selectors, or can it also patch waits/assertions when latency spikes? If I run tests on a private staging env behind VPN, will the...

This is brilliant—turning a multi-week email marathon into a 10-minute conversation is exactly what distributed teams needed. Love that the AI keeps context when you pivot from “beachfront villa” to “actually, give me mountain cabins with fiber internet” mid-thread. Quick ask: does the copilot surface visa/entry requirements and seasonal weather averages alongside venue recs? That would save...