Senior QA Tester — AI Product UAT (Web, ~40-50 hours, fully scoped plan provided)

I'm looking for a Top Rated Plus QA tester for a structured UAT cycle on Juliet, an AI marketing product. The full test plan is built — 156 test cases across 4 projects, each project ~9-10 hours of work. You execute the plan, log bugs in Linear, and hand off a recorded session. What you'll test Juliet (https://agentic-dev.juliet.space/) is an AI builder where users chat with it to generate marketing websites and assets. The UAT runs the product through 4 realistic agency client scenarios: A HIIT gym chain (reference image → site) A developer SaaS launch (URL extract → site) A DTC nutrition brand (PDF brief → site) A home services franchise (reference image → site) Each project tests: brand extraction, site building, Supabase backend, Stripe checkout, Resend email, Google Sheets append, AI features (OpenAI + pgvector in one project), and 4 "torture moments" per project that test state persistence (refresh, tab close, asset history restore, multi-step undo). What's prepared for you: 156 test cases with exact prompts to paste, exact files to upload, exact expected results 15 upload files pre-staged in a single folder Spreadsheet access for bug logging Stripe test cards / test-mode keys Drive folder for recordings + screenshots Please review actual flow here: https://docs.google.com/spreadsheets/d/1qJUJrGQYp2r3bA7C3QhVApnSuFskI97j/edit?usp=sharing&ouid=114987900721998126290&rtpof=true&sd=true What I need from you: Chrome + DevTools (Console + Network) open at all times Screen recorder running the entire session (Loom / OBS) Screenshot at every confusion / failure Console errors logged per case Bugs filed in Linear with case ID + reproduction steps + recording timestamp Never skip a torture moment, even if the previous one passed Mac/PC with Chrome and good memory required (test plan assumes ~/uat-upload-kit/ folder) Critical skills Strong DevTools — you should be able to spot if sk_... or OPENAI_API_KEY leaks in a network request Supabase familiarity (verify table rows, RLS policies, Edge Function triggers) Stripe test mode (verify webhook fires, checkout flow, dashboard reconciliation) JSON-LD / schema.org validation Comfortable with a "frantic agency marketer" persona (you paste prompts that escalate in tone — copy them exactly, don't paraphrase) Honest reporting: if you can't judge whether a brand voice matches a brief, mark the case "Executed — needs founder review" rather than guessing Pass. This is more valuable than a clean sheet. Budget + timeline $5-8/hr, est. 40-50 hours Start within 1 days of hire Deliver in 3-5 calendar days Bonus available for high-quality bug write-ups (clear repro + recording timestamps) To apply, answer these 2 questions (skip generic cover letters — I won't read them) Have you tested an AI / LLM product before? Name it. If not, that's OK — say so. What's the last bug you found by inspecting the DevTools Network tab? One sentence.