Thinking Machines Lab
Research, Tinker, RL Systems
$350K–$475K
San Francisco · On-site · Full-time · Posted Oct 19, 2026
- Education
- Bachelor’s degree or equivalent experience; a PhD or equivalent industry research experience is preferred.
- Visa sponsorship
- Visa sponsorship is available; success is not guaranteed for every candidate or role.
About the job
Build training systems and post-training techniques for Tinker, a fine-tuning API, spanning reinforcement learning algorithms, data pipelines, numerics, and kernels.
What you'll do
- Develop frontier customization techniques and improve Tinker’s post-training engine.
- Co-design RL algorithms and training systems with internal researchers and external partners.
- Debug RL runs, optimize post-training pipelines, and help users achieve frontier-level results.
What they ask for
- Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline, with strong theoretical and empirical grounding.
- Proficiency in Python and familiarity with deep learning frameworks such as PyTorch, TensorFlow, or JAX.
- Comfort debugging distributed training and writing scalable code.
- Clear written communication and strong interest in working on Tinker and increasing its usefulness and adoption.
Nice to have
- Experience with RL training stability techniques for large runs, low-precision training and inference, or scaling studies for large models.
- Hands-on experience with LLM serving stacks or contributions to open-source training or inference frameworks; PhD or equivalent industry research experience.
Pay
$350,000 - $475,000 USD annual salary range (base salary)
Tools
Python, PyTorch, TensorFlow, JAX, SGLang, vLLM, TokenSpeed
AI focus
Develop RL algorithms, training systems, and post-training techniques that enable users to customize and post-train frontier models.
Model-written summary · the original posting is authoritative