Research, Coding Agents
$350k - $475k • San Francisco • FullTime
Posted 1mo ago
Job Location
San Francisco
Tech Stack
Remote Work Policy
On-site
Employment Type
FullTime
Categories
AI Agent Engineer
About the job
The Coding Agents team at Thinking Machines is dedicated to enhancing our AI models' capabilities in agentic coding. This involves developing expertise in writing, debugging, and reasoning about code across complex, multi-turn tasks. As a member of this small, high-leverage team, you will be instrumental in defining the recipes, data, and infrastructure that drive coding proficiency improvements in every model release. You will own the entire post-training stack for coding capabilities, encompassing synthetic and human data generation, RL environments, reward design, and large-scale training runs. This research-focused role offers significant ownership, allowing you to shape the technical direction of our coding agent development.
Responsibilities
- Design and execute RL training jobs focused on agentic coding, refining recipes and data.
- Develop and enhance sandboxed coding environments and reward signals for model training and evaluation.
- Generate and curate high-quality synthetic coding data, and build scalable data pipelines.
- Design evaluations to measure real-world coding usefulness and train models to improve usability.
- Debug and analyze large RL runs to identify and address issues like confounders and reward hacking.
- Collaborate with infrastructure, evaluation, and post-training teams to integrate improvements into model releases.
Requirements
- Strong engineering skills with the ability to contribute code and debug complex codebases.
- Proficiency in Python and at least one deep learning framework (e.g., PyTorch, TensorFlow, JAX).
- Experience with debugging distributed training and writing scalable code.
- Bachelor's degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline.
- Clear communication skills, with the ability to explain complex technical concepts in writing.
- Experience building synthetic data pipelines and systems adopted by others.
- Experience owning the end-to-end cycle of identifying and closing model usability gaps.
- Experience making large-scale agentic RL infrastructure reliable.
- Experience improving the coding capabilities of a frontier model.
- PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline, or equivalent industry research experience.
Benefits
- Generous health, dental, and vision benefits
- Unlimited PTO
- Paid parental leave
- Relocation support