Autonomy Engineer Intern, Reinforcement Learning at Skydio

Skydio is a US drone company and a leader in autonomous flight, making drones, docks and software for inspection, public safety and defence. As a reinforcement learning intern you develop and deploy learned policies that help its aircraft plan, navigate and control themselves more safely and efficiently, and test them on real drones.

Key responsibilities

• Train navigation and obstacle-avoidance policies for cluttered 3D scenes
• Combine learned value functions with trajectory optimisation
• Build sim-to-real training with Isaac Lab and domain randomisation
• Learn assistive policies for shared human and autonomy control
• Explore multi-agent coordination

Requirements

• PhD student in robotics, machine learning, controls or similar
• Strong RL, control theory and motion planning
• Python (PyTorch, JAX or RLlib) and C++ or CUDA
• Robotics simulation and sim-to-real experience
• Experience deploying policies on real robots or vehicles

Nice to have

• Publications at CoRL, ICRA, IROS, RSS or NeurIPS
• Onboard inference optimisation

Location and schedule

• Skydio HQ, San Mateo, California; onsite
• 2027

Salary and benefits

USD 47 an hour for undergraduates, USD 53 for graduate students and USD 70 for PhD students, depending on skills and experience.

How to apply

Apply on the company's own site.