Skydio
Autonomy Engineer Intern, Reinforcement Learning, 2027 (HQ)
San Mateo, California, United States · On-site · Internship · Posted Oct 26, 2026
- Education
- PhD student in Robotics, Machine Learning, Controls, or related field.
- Visa sponsorship
- For positions located in the United States of America, Skydio, Inc. uses E-Verify to confirm employment eligibility.
About the job
Develop and deploy reinforcement-learning and related policies that enable Skydio drones to navigate, plan, and control themselves safely and reliably, validating the work in simulation and on real aircraft.
What you'll do
- Train navigation, obstacle-avoidance, shared-control, and multi-agent policies for drone missions.
- Combine learned value functions or cost shaping with trajectory optimization and safety constraints.
- Build simulation-to-real training workflows and test policies on production-ready drones.
What they ask for
- PhD student in Robotics, Machine Learning, Controls, or a related field.
- Strong fundamentals in reinforcement learning, control theory, and motion planning; comfort with safety and robustness concepts.
- Proficiency in Python and at least one of C++ or CUDA.
- Hands-on robotics simulation and sim-to-real experience; experience training or deploying policies on real robots or autonomous vehicles.
Nice to have
- Publications in CoRL, ICRA, IROS, RSS, or NeurIPS.
- Experience with onboard inference optimization, modern policy-learning methods, multi-agent reinforcement learning, or distributed training.
Pay
$47/hr (hourly rate for Undergrad); $53/hr (hourly rate for Masters); $70/hr (hourly rate for PhD students)
Tools
Python, PyTorch, JAX, Ray RLlib, C++, CUDA, Isaac Lab, MuJoCo
AI focus
Develop reinforcement-learning and adjacent policy-learning methods for autonomous drone navigation, planning, control, shared control, and multi-agent coordination; train in simulation and validate on real aircraft.
Model-written summary · the original posting is authoritative