Autonomy ML Engineer

About the role

SWARM is building the intelligence layer for autonomous drones and seeks an Autonomy ML Engineer to create learning‑based flight agents. You will work on perception, decision‑making and control in simulation and progressively on real aircraft, improving robustness, safety and performance in unseen environments.

Key responsibilities

• Train autonomous flight policies for navigation, obstacle avoidance and mission execution.
• Design observations, action spaces, reward functions, curricula and evaluation strategies.
• Run simulation experiments, analyse failures and develop systematic improvements.
• Enhance robustness, safety, speed and generalisation of learned policies.
• Combine machine‑learning approaches with classical planning and control where appropriate.
• Transfer successful policies from simulation to real drones.
• Develop perception and autonomous decision‑making capabilities for commercial projects.
• Work independently on open‑ended technical problems from analysis to experimental validation.

Required profile

• 2–4 years of relevant experience in autonomy, reinforcement learning or robotics, preferably in a professional or research setting.
• Hands‑on experience building and evaluating reinforcement‑learning systems that can be iteratively improved.
• Ability to design reproducible experiments and work autonomously on open‑ended problems.

Required skills

• Python
• PyTorch
• Reinforcement learning (continuous control)
• Algorithms such as PPO, SAC, TD3
• Robotics fundamentals and physics simulation
• Simulation tools: Isaac Sim, Isaac Lab, MuJoCo, PyBullet, Gazebo
• Computer vision and depth‑based perception
• Curriculum learning, domain randomisation, sim‑to‑real transfer
• Multi‑agent reinforcement learning
• Motion planning