Research, Finetuning Science
$350k - $475k • San Francisco • FullTime
Posted 1mo ago
Job Location
San Francisco
Tech Stack
Remote Work Policy
On-site
Employment Type
FullTime
Categories
AI Research Engineer
About the job
At Thinking Machines, we build tools that empower individuals to customize AI models for their specific needs, including the ability to train model weights. In this role, you will focus on frontier customization techniques and contribute to developing a leading post-training engine called Tinker. Your work will involve a comprehensive understanding of Reinforcement Learning (RL) science, and your findings will directly influence Tinker's training defaults, API design, and the open-source Tinker Cookbook. You will collaborate with internal research teams and contribute to open science initiatives for external partners.
Responsibilities
- Advance the science of fine-tuning and frontier post-training techniques.
- Contribute to areas like LoRA and parameter-efficient fine-tuning to enhance customization quality, efficiency, and reliability.
- Ship research into product by informing Tinker's training defaults and primitives, and codifying best practices in the Tinker Cookbook.
- Improve the stability, efficiency, and reliability of large-scale fine-tuning and RL runs on Tinker.
- Share learnings through papers, technical blog posts, and community contributions.
Requirements
- 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 (e.g., PyTorch, TensorFlow, or JAX).
- Comfort debugging distributed training and writing scalable code.
- Clarity in communication and ability to explain complex technical concepts in writing.
- Strong interest in enabling custom AI models.
- A strong grasp of probability, statistics, and ML fundamentals.
- Experience managing or analyzing human data collection campaigns or large-scale annotation workflows.
- Experience with RL training stability techniques for large runs.
- PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or equivalent industry research experience.
Benefits
- Generous health, dental, and vision benefits
- Unlimited PTO
- Paid parental leave
- Relocation support