Member of Technical Staff, Evals

What you'll do

• Design evaluation frameworks and benchmarks for LLMs, agents, multimodal models, and reinforcement-learning environments.
• Develop reward models, verifiers, graders, and human-data quality systems; research reliability and resistance to reward hacking.
• Run evaluation and research iterations, and contribute benchmarks, software, research, and technical writing publicly.

What they ask for

• PhD in ML/AI, computer science, data science, or a related field, or equivalent research experience in industry.
• Publications at top AI/ML venues such as NeurIPS, ICML, ICLR, or COLM.
• Strong Python skills; experience building scalable software and working with agents.
• Strong knowledge of frontier AI, including benchmarks, evaluation techniques, agent harnesses, post-training recipes, and data shapes.

Nice to have

• Strong GitHub profile or open-source contributions, or a product portfolio.
• Comfort operating in an ambiguous, fast-moving environment with substantial ownership.

Pay

USD $200K – $350K base salary

Tools

Python, LLMs

AI focus

Develops evaluations, benchmarks, reward and verifier systems, agent-evaluation methods, and data-quality techniques for frontier AI.