Autonomy Engineer Intern, Reinforcement Learning, 2027 (Zurich)
Core
Develop and deploy reinforcement learning policies to enable autonomous drones to plan, navigate, and control themselves in complex environments.
Role type
PhD-level research intern (reinforcement learning for robotics)
Builds
Autonomous flight software for inspection, public safety, and defense drones
Domain
Robotics, Autonomous Systems, Reinforcement Learning
Deliverable
production ML models
Required skills
Reinforcement learning, Control theory, Motion planning, Python, C++ or CUDA, Robotics simulation (Isaac Lab/MuJoCo/PyBullet), Sim-to-real transfer
Preferred skills
Publications in top robotics/ML conferences, Onboard inference optimization, Modern policy learning (diffusion, IL/BC, offline RL), Multi-agent RL
Technologies
PyTorch, JAX, Ray RLlib, Isaac Lab, TensorRT
Responsibilities
Train policies for navigation and obstacle avoidance in cluttered 3D scenes, Fuse learned cost shaping with trajectory optimization, Build scalable datasets and training loops with domain randomization, Develop assistive policies for human-in-the-loop shared control, Explore decentralized coordination for multi-agent fleet workflows
Seniority
Intern, Research
