Research Engineer - Agency and Reasoning
Core
Conduct novel research in reinforcement learning, post-training, and human preference learning to apply ideas at scale to next-generation language models.
Role type
Research Engineer (Agency and Reasoning)
Builds
Next generation of language models
Domain
Artificial Intelligence / Large Language Models
Deliverable
production ML models
Required skills
reinforcement learning, language-model-supervised fine-tuning, preference-learning methods (DPO, simPO), context-length extension, iterative fine-tuning, data engineering, synthetic data generation, PyTorch, Python
Preferred skills
rapid prototyping, strong research taste, ability to execute projects from conception to write-up
Technologies
PyTorch, Python
Responsibilities
Perform novel research in RL and post-training; apply research ideas at scale to language models; engage in data engineering and synthetic data generation; iterate on model behaviors through fine-tuning
Seniority
Mid-Senior, hands-on IC
