Quick overview
Job description
About the Role
Join a small applied research team working with frontier AI labs on challenging post-training problems. This research-first role focuses on reinforcement learning, training data, and experiments that help improve model capabilities.
What You'll Do
Work with research teams to translate open-ended goals into concrete environments, datasets, experiments, and deliverables.
Design and run post-training workflows, including reinforcement learning methods such as GRPO, preference optimization, and supervised fine-tuning.
Shape and inspect high-fidelity trajectories and training data, with attention to correctness, edge cases, and quality.
Build reward functions, rubrics, validators, and evaluation harnesses that resist reward hacking and measure capability gains.
Analyze post-training signals and experiment results across distributed GPU clusters to guide subsequent research.
What We're Looking For
At least 2 years of experience as an applied researcher or research scientist, with hands-on model training experience.
Practical reinforcement learning and agent experience beyond standard machine learning engineering work.
Research experience at a frontier AI lab, AI research startup, or AI data lab, or evidence of impactful research through a publication or a model performing well on a notable benchmark.
Strong knowledge of reinforcement learning and training data, supported by research or applied work.
Experience with Python, PyTorch, FSDP, FastAPI, GRPO, and distributed GPU clusters.
A PhD involving reinforcement learning research or experience at a frontier AI lab is relevant background.
Compensation & Benefits
Base salary range is $300,000 to $500,000 annually. Visa sponsorship is available.
Location
Hybrid in New York, New York, with an additional location in Seattle, Washington.