Meridial

Private Company
ShareTweet

Open Positions27

World Wide - RemoteContractArtificial IntelligencePosted
  • Design and advise on rubric architecture used to evaluate the quality of AI outputs.
  • Create, edit, and certify 'Golden Tasks' that serve as ground truth for model training.
  • Analyze model failures such as hallucinations and logical fallacies.
  • Provide technical breakdowns explaining why models fail and how experts would resolve issues.
  • Consult on the difficulty and nuance of prompts to ensure rigorous AI testing.
Showing 1 of 27 positions