AI Research Scientist, Reinforcement Learning (LLM) and Post-Training
Core
Invent and analyze reinforcement learning algorithms to advance post-training and interactive learning for large generative models applied to engineering and hardware-adjacent tasks.
Role type
Senior IC AI Research Scientist (Reinforcement Learning)
Builds
Post-training LLMs and code models for structured engineering tasks
Domain
AI Research / Reinforcement Learning / Large Language Models
Deliverable
production ML models
Required skills
Reinforcement learning theory, policy optimization, reward modeling, empirical study design, distributed training, LLM post-training, preference optimization
Preferred skills
Publication record in RL, training at non-trivial scale, familiarity with compilers/kernels/EDA workflows
Technologies
GPUs, distributed jobs, NeurIPS/ICML/ICLR venues
Responsibilities
Research and develop RL methods for post-training LLMs and code models; Design reward models, curricula, and training recipes; Characterize failure modes and propose mitigations; Collaborate with infra engineers to scale training; Publish at top venues and lead internal RL roadmap
