production workflows · data pipelines
ABOUT THE ROLE
Join an early-stage AI company building specialized models and learning systems for production agents. You will work across research and production engineering to improve how models are trained, evaluated, and deployed.
WHAT YOU'LL DO
- Build evaluation suites, graders, and data pipelines using production workflows and traces.
- Own post-training projects in customer settings, from data preparation and evaluations to reinforcement learning environments and training runs.
- Develop agent environments and harnesses for reproducible experiments and training data generation.
- Run and analyze post-training experiments, reproduce promising research, and debug related systems.
- Turn successful experiments into reliable components of a post-training and continual learning platform.
WHAT WE'RE LOOKING FOR
- One to five years of engineering experience at an AI startup or in an ML research group, with hands-on work on large language models.
- Experience training, evaluating, and serving LLMs, beyond using model APIs or prompt and agent wrappers.
- Strong Python skills and foundations in machine learning and computer science.
- Hands-on experience with PyTorch, LLM training, evaluation harnesses, or model serving.
- A substantive project, research contribution, shipped system, internship, or open-source work demonstrating technical depth and ownership.
- Interest in post-training or inference, demonstrated through experiments, research, open-source work, or startup experience.
- Ability to learn unfamiliar systems, debug independently, comm