Staff Applied Research Engineer

Applied research role focused on enabling continuous learning in production for AI agents.

• •The OpenPipe team at CoreWeave is building tools to help agents learn from experience, a critical step towards enabling autonomous, long-task execution.
• •This role focuses on systematically identifying and solving major bottlenecks in agent development.
• •Key Responsibilities Generate and investigate research ideas towards solving obstacles to continuous learning in production.
• •Validate research directions across real customer tasks.
• •Work with the broader OpenPipe team on developing tools for agents to learn from experience.
• •Contribute to the development of novel training methods and systems for LLMs.
• •Tackle challenges in building simulated environments and improving data efficiency in training methods.
• •Requirements Trained LLMs to be SOTA on specific tasks.
• •Opinions on sequence-level vs. token-level importance ratios.
• •Experience with LLM post-training, including supervised fine-tuning, reinforcement learning, reward modeling, and policy optimization.