Senior Research Scientist LLM
Core
Designing and running post-training experiments to teach LLMs to reason, use tools, and follow intent.
Role type
Senior Research Scientist (LLM Post-Training)
Builds
Production LLMs with improved reasoning, controllability, and alignment
Domain
Generative AI / Large Language Models
Deliverable
production ML models
Required skills
Post-training techniques (DPO, GRPO, SFT, rejection sampling), Reinforcement Learning, Reward modeling, Evaluation framework design, Data pipeline construction, End-to-end project ownership
Preferred skills
Experience with open-source models, Commercial-scale deployment experience
Technologies
DPO, GRPO, SFT, Rejection sampling
Responsibilities
Design and run post-training experiments, Build and scale RL and post-training infrastructure, Develop reward models and evaluation frameworks, Work with vendors to build high-quality datasets, Own end-to-end projects from data collection to model improvement
