Research Engineer, Benchmarking

Refresh AI; San Francisco (CA); On-site; USD 120,000 - 150,000; Full time

Job summary

Refresh AI is seeking a Research Engineer in San Francisco to push the boundaries of benchmarking technology. You will build benchmarks that labs use for evaluating coding abilities and computer-use capability. Your role will require expertise in reinforcement learning and supervised fine-tuning, as well as a willingness to engage in full-stack development on our core tech stack (Vercel, Supabase, Render). This position offers a full-time role in a dynamic environment.

Qualifications

• Experience with reinforcement learning and supervised fine-tuning.
• Ability to ship full-stack work on core technologies when required.

Responsibilities

• Build benchmarks for coding and computer-use capability.
• Translate expert workflows into rigorous evaluations.
• Run evaluations against frontier models and publish verifiable numbers.

Skills

Reinforcement learning

Supervised fine-tuning

Full-stack development

Tools

Vercel

Supabase

Render

Job description

# Research Engineer, BenchmarkingEngineeringSan FranciscoFull-timeBuild the benchmarks frontier labs use to measure real-world coding and computer-use capability. Translate expert workflows into rigorous, verifiable evaluations, run them against frontier models, and publish numbers that hold up under adversarial scrutiny.Across every role at Refresh: you're willing to ship full-stack work on our core stack (Vercel, Supabase, Render) when it's needed, and you're comfortable with reinforcement learning and supervised fine-tuning at a high level.