Una candidatura completa en un minuto
Currículum y carta de presentación personalizados, listos para enviar.
San Francisco Data is building training environments for frontier models using production traces. Our tight-knit SF team operates five days a week from the office, delivering production-ready environments and benchmarks.
You will join engineers and researchers to extend training environments, analyze agent behavior, and run rigorous experiments. The role rewards independent problem solving and hands-on ML pipeline work in a fast-moving startup setting.
San Francisco Data turns production traces into training environments for frontier models. We’ve reached $100M+ in annualized revenue run rate and raised a $30M seed round led by Khosla Ventures.
Our founders previously grew Mercor’s code team from $0 to $500M in annualized revenue run rate, leading its largest code engagements and synthetics work. Our team brings experience from Google DeepMind, Stanford SAIL, and Berkeley RISE Lab, including TerminalBench co-authorship.
We’re a small team of highly technical engineers working out of our SF office 5 days a week.
Staffed on the same team as our full time engineers and researchers, working on developing new training environments and task formats for frontier models using production agent traces.
Analyze agent behavior and failure cases, and run experiments on new benchmarks and task formats
Develop tools to generate, test, and evaluate training tasks, helping improve the quality and reliability of our environments.
Prior research experience, open-source contributions, or work on agent benchmarks and ML pipelines are a plus.
Hands-on experience building AI agents, reinforcement learning environments, synthetic data pipelines, or evaluation benchmarks, with a strong understanding of the challenges involved in training and evaluating frontier models.
Proven ability to independently tackle complex technical problems, build robust systems, and design rigorous experiments to evaluate performance and drive improvements.
Curiosity about how AI systems learn and fail, and an interest in building more effective ways to train and evaluate them.
We care more about what you’ve built, how you think, and your willingness to learn than specific credentials or publications.