Research Scientist, RL Training
Redwood City, California, United States of America
Full Time
2 days ago
$200,000 - $275,000 USD
Visa Sponsor
Key skills
CloudNode.jsPythonPyTorchAIMachine LearningMLLLMLarge Language Models
About this role
Role Overview
- Research and implement reinforcement learning techniques including GRPO, RLHF, RLAIF, DPO, and reward modeling
- Design and build data pipelines that generate high-quality training signal for RL workflows
- Prototype and iterate on end-to-end RL training recipes
- Work closely with research scientists, ML engineers, and delivery teams
- Stay current with the latest developments in large-scale muli-node LLM training
Requirements
- Deep expertise in reinforcement learning from human or AI feedback
- Experience training or fine-tuning 30B+ large language models at scale
- Strong proficiency in Python and ML frameworks, especially PyTorch and HuggingFace
- Solid software engineering fundamentals
- Familiarity with ML infrastructure and cloud platforms
- Comfort operating in a high-iteration environment
- Ph.D. in machine learning, reinforcement learning, or a related field strongly preferred
Tech Stack
- Cloud
- Node.js
- Python
- PyTorch
Benefits
- Health insurance
- Professional development opportunities
- Flexible work arrangements