Product.ai / Projects / Production Eval Harness — error-analysis-first failure taxonomy on Product.ai chat or Alloy / Apply
Apply

Production Eval Harness — error-analysis-first failure taxonomy on Product.ai chat or Alloy

Build a production eval harness for one Product.ai surface (chat or Alloy). The candidate picks the surface, reviews 50-100 real production traces, builds a failure taxonomy via open-coding then axial-coding, ships 20-50 task evals derived from real failures, wires them into a CI gate, and instruments a continuous-production-eval sampling pipeline with privacy-respecting trace capture. Output is a working eval harness producing real scores on real traffic, not a deck.

STEP 1 — APPLY

Apply: Production Eval Harness — error-analysis-first failure taxonomy on Product.ai chat or Alloy

Submit your details below. After this step you'll record a short video questionnaire.

PDF, DOC, or DOCX · max 50MB

A headshot for your candidate card — JPG, PNG, or WEBP · max 8MB. Optional.

A piece of writing or analysis that shows how you think — process doc, system design, deep research. PDF, DOC, or DOCX · max 50MB.

FIT & LOGISTICS

Most of our roles are on-site or hybrid in Santa Monica. We need your own answers here — please don't rely on what's on your LinkedIn.

On-site / hybrid / remote *
Employment type

Paid trials are contract by default. Pick "Full-time" if you're hoping the trial converts.

By submitting, you agree to our Privacy Policy. We use this information to process your application.

After submitting, you'll be taken to the Hireflix video questionnaire.

If you encounter any issues submitting your application, please reach out to hiring@product.ai.