About this role
Research Scientist - RL Training at Snorkel AI. Location: Redwood City or San Francisco or United States. Role: researching RL, designing pipelines, prototyping recipes Requirements: Deep RL and reward-modeling expertise; experience training/fine-tuning large LLMs; proficiency in Python, PyTorch, HuggingFace; familiarity with distributed training and cloud infrastructure; PhD preferred. Category: Research and Development (R&D) Seniority: Senior Level Tools: Python, PyTorch, HuggingFace, Verl, SkyRL, AWS, GCP, Kubernetes, Slurm, Terminal-Bench Commitment: Full Time Workplace: Hybrid Languages: English