Research Engineer

About this role

Join an AI infrastructure company building systems for training and evaluating frontier AI agents. As a Research Engineer, you'll design benchmarks, synthetic data pipelines, and quality control automation while working alongside world-class researchers in a high-ownership, early-stage environment.

What you'll do

• Build systems for creating training environments, improving data quality, and converting real-world workflows into tasks and benchmarks
• Design experiments to understand model behavior, agent failure modes, and identify data quality issues
• Develop tools for researchers and data vendors to create higher-quality tasks, trajectories, and feedback loops
• Manage the full lifecycle of agent training data from task design through trajectory collection, evaluation, and validation
• Partner with external vendors to identify bottlenecks and improve data engine quality and throughput
• Build metrics and analyses to assess task and environment usefulness for training frontier agents

What they're looking for

• Python
• Docker
• Linux
• Benchmark design and evaluation
• Data pipeline architecture
• Metrics and validation methodology
• Reinforcement learning (preferred)
• Independent infrastructure building

Benefits

• Competitive salary ($150K–$250K annually)
• Equity participation in well-resourced AI company
• Visa sponsorship available
• Work with International Olympiad medalists and top-tier researchers
• High ownership and direct impact on frontier AI development
• On-site collaborative environment in San Francisco