Research Engineer - Data and Evaluation
Core
Design and build challenging evaluations, executable environments, and RL data pipelines for training frontier AI models and agents.
Role type
Senior IC research engineer (RL evaluation & environments)
Builds
RL training data pipelines, evaluation systems, and executable agent environments
Domain
Frontier AI / Reinforcement Learning / Agent Systems
Deliverable
production ML models
Required skills
Python, data pipeline engineering, RL workflows, model evaluation design, failure mode analysis, reward hacking detection, agent environment construction
Preferred skills
RL post-training data, synthetic data generation, statistical testing, contamination detection, Verifiers framework experience
Technologies
Python, Verifiers, RL frameworks
Responsibilities
Design challenging evaluations for coding and long-horizon agents; Build executable environments with realistic task setup; Develop scoring logic and verifiers to detect reward hacks; Trace model failures to distinguish limitations from environment issues; Build and scale RL data pipelines for task generation; Collaborate with researchers to validate training data improvements
Seniority
Senior, hands-on IC
