The work
We are testing AI agents on realistic shopping and delivery tasks. You'll review AI-drafted tasks (prompt, grading rubric) along with model attempts, and decide whether each failure is fair:
- Was the information available?
- Does the rubric accurately grade what the prompt asked for?
- Would a valid answer have been marked wrong?
- Is the rubric structure correct per the guidelines?
Fix the criteria that aren't fair, flag the tasks that can't be saved, and write a short note on each call.
Must-haves
- Substantial experience writing or reviewing rubrics for LLM evals. Please list the projects in your application.
- Can read a long agent transcript and find where it went wrong.
- Precise writing, and you check the data instead of trusting the reference answer.