All interviews
Cohere logo

Cohere

Mid

Evaluate and debug AI coding-agent outputs to sharpen Cohere's models

Cohere is hiring a part-time independent contractor (16+ hrs/week, Canada-based, 12-month contract) to review and debug code, navigate repository architecture, and analyze coding-agent trajectories for model training and evaluation. The role wants 3-5 years of software engineering experience with proficiency across Python plus Java, JavaScript, Go, and SQL, prior API design/build/deploy experience, and ideally hands-on exposure to code agents like Claude Code, Cursor, Codex, or OpenCode — this is evaluation/annotation work, not product engineering.

Practice this interview

Free · a live voice mock calibrated to this exact role

Start the mock interview

Likely format

Initial resume/writing-sample screening, then a virtual annotation test (coding take-home + writing sample), then a video screen with the Operations team, then offer.

What this interview tests

  • Multi-language code review (Python, Java, JavaScript, Go, SQL)
  • Debugging and evaluating AI-generated code / agent trajectories
  • API design critique
  • Repository architecture navigation
  • Precision and rigor in labeling/evaluation judgments

Common question themes

Review this code snippet or diff and identify what's wrong with it

Walk through evaluating whether a coding agent's trajectory actually solved the task correctly

Critique an API design — what's missing or wrong given the stated requirements

Describe a time you caught a subtle bug that looked correct on the surface

How do you navigate an unfamiliar codebase quickly to assess whether a change is sound

What's your experience using coding agents like Claude Code, Cursor, or Codex, and where have you seen them fail

View the original posting

All Cohere Data Annotation Specialist interviews

All Cohere interviews

Related interviews