Research, Tinker, RL Systems

About the job

Responsibilities

• Develop frontier customization techniques and build a post-training engine.
• Engage with researchers and companies using Tinker to understand their needs and push the platform's limits.
• Contribute to open science and collaborate with external partners.
• Co-design RL algorithms and training systems across the entire stack, from science to numerics and kernels.
• Debug RL runs, optimize post-training pipelines, and help users achieve frontier-level results.

Requirements

• Bachelor's degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline.
• Proficiency in Python and familiarity with deep learning frameworks like PyTorch, TensorFlow, or JAX.
• Comfort debugging distributed training and writing scalable code.
• Ability to clearly explain complex technical concepts in writing.
• Strong interest in working on Tinker and increasing its usefulness and adoption.
• Strong grasp of probability, statistics, and ML fundamentals.
• Experience with RL training stability techniques for large runs.
• Familiarity with low-precision training and inference (numerics, quantization).
• Hands-on work with LLM serving stacks (e.g., SGLang, vLLM).
• Experience with scaling studies for large models.
• Contributions to open-source training or inference frameworks.
• PhD in a relevant discipline or equivalent industry research experience.

Benefits

• Health, dental, and vision benefits
• Unlimited PTO
• Paid parental leave
• Relocation support