Principal AI Research Scientist

Join an early-stage AI research team developing systems that learn and improve while supporting engineering work in the physical world. As a principal-level individual contributor on a team of AI researchers and engineers, you will set research direction and bring new ideas from experimentation into production.

• Lead research from initial exploration through implementation in continual learning, fast adaptation, self-improvement, and memory formation.
• Set technical direction and refine the research roadmap in collaboration with AI researchers and engineers.
• Take research into production, working across data generation, evaluation, and product integration.
• Develop and test approaches that improve system capabilities, and publish research results.

• A PhD or equivalent experience in computer science, robotics, engineering, mathematics, physics, or a related field, with 3 to 15 years of AI research experience.
• Applied or production research experience, including taking research ideas into working systems, preferably in an early-stage environment.
• Research depth in continual or lifelong learning, fast or test-time adaptation, self-improving systems, memory, meta-learning, knowledge retention, transfer learning, reinforcement learning, or related areas.
• Hands-on post-training experience with supervised fine-tuning and reinforcement learning methods, such as RLVR, GRPO, or PPO.
• A publication record in leading AI or machine learning venues, with experience in agentic or long-horizon systems a plus.
• Fluency in Python, PyTorch or JAX, and distributed training frameworks.

Base salary is $200,000 to $400,000 USD annually. Visa sponsorship is not available.

This is an on-site role in San Mateo, California.