Reinforcement Learning Planning Research Intern

About the role#

This internship focuses on the development of a Safety-Critical Trajectory Correction (STC) module. As a real-time safety overlay, the STC acts as a fallback mechanism that intercepts and minimally perturbs intended trajectories when collision risks are detected. You will design, train, and validate this architecture using Deep Reinforcement Learning to provide a continuous, constrained safety barrier for the autonomous vehicle fleet.

What you'll do#

• Conduct research on reinforcement learning to generate safe trajectories for autonomous driving, with the goal of achieving publishable results.
• Develop and benchmark deep learning techniques.
• Collaborate with team members to optimize and integrate developed techniques into the production perception and autonomous vehicle stack.

What you'll need#

• Currently pursuing an MS or PhD in Computer Science, Electrical Engineering, Mathematics, Statistics, or a related field.
• A thorough understanding of reinforcement learning.
• 1-2 years of experience implementing and training models in at least one deep learning framework, such as PyTorch, TensorFlow, or Jax.

Location & details#

• Location: Santa Clara, California, USA.
• Term: Summer 2026.
• Modality: On-site, full-time.
• Compensation: This is a paid internship.