Member of Technical Staff, Coding Agents

What you'll do

• Create and publish coding benchmarks that challenge state-of-the-art agents.
• Study the frontier of coding agents and publish impactful findings.
• Help AI labs improve coding models through benchmarks, open-source tools, research, and technical writing.

What they ask for

• Passion for building software with agents and regular experimentation with coding agents.
• Comfort working in an ambiguous, fast-moving environment with substantial ownership.
• At least one of: experience working on a coding benchmark; published AI-for-software or code-generation research; extensive software engineering experience or strong open-source GitHub contributions, plus deep knowledge of agent harnesses.

Pay

USD $200K – $350K (base salary)

AI focus

Develop coding-agent benchmarks, reward and verifier systems, evaluation methodologies, and data-quality techniques; investigate agent capabilities and help improve coding models.