Senior Autonomy Machine Learning Engineer (Reinforcement Learning & Autonomous Systems)

About the Role

At SWARM, we are building the intelligence layer for autonomous drones.

We are looking for an Autonomy ML Engineer to develop learning-based flight agents capable of perceiving, deciding, and acting in simulation and, progressively, on real aircraft.

The main objective of the role is to improve the performance of our autonomous flight systems in unseen environments, increase their robustness and safety, and transfer those capabilities into real-world applications such as inspection, surveillance, and other autonomous missions.

This is not a generative AI, LLM, or prompt-engineering role. We are looking for hands-on experience in reinforcement learning, robotics, simulation, and autonomous systems.

The position is based in Andorra and is part of SWARM’s AI / Autonomy team

Responsibilities

• Train autonomous flight policies for navigation, obstacle avoidance, and mission execution.
• Improve agent performance in new environments, not only in scenarios used during training.
• Design observations, action spaces, reward functions, curricula, and evaluation strategies.
• Run simulation experiments, analyze failures, and develop improvements systematically.
• Improve the robustness, safety, speed, and generalization of learned policies.
• Develop capabilities for increasingly complex missions, including search and rescue, interception, and multi-agent coordination.
• Combine machine learning approaches with classical planning and control where appropriate.
• Contribute to transferring successful policies from simulation to real drones.
• Develop perception and autonomous decision-making capabilities applicable to commercial projects.
• Work with a high degree of autonomy on open-ended technical problems, from initial analysis through experimental validation

Requirements

We are looking for someone with approximately 2–4 years of relevant experience, ideally in a professional or research environment related to autonomy, reinforcement learning, or robotics. We care especially about what you have built and demonstrated in practice

Core Requirements

• Strong proficiency in Python and PyTorch.
• Hands-on experience with reinforcement learning and continuous control.
• Knowledge of algorithms such as PPO, SAC, TD3, or related methods.
• Experience with robotics or physics simulation.
• Experience with reward design, policy evaluation, and failure analysis.
• Ability to design structured and reproducible experiments.
• Understanding of robotics fundamentals: position, velocity, orientation, and coordinate frames.
• Ability to work independently on open-ended technical problems.
• Experience training autonomous or reinforcement-learning systems that perform consistently and can be improved through systematic experimentation

Especially Valuable Experience

• Experience with Isaac Sim / Isaac Lab, MuJoCo, PyBullet, Gazebo, or similar tools.
• Computer vision, depth-based perception, or learned perception.
• Curriculum learning and domain randomization.
• Sim-to-real transfer.
• Multi-agent reinforcement learning.
• Motion planning and control.
• ROS2, PX4, or MAVLink.
• ONNX, TensorRT, edge deployment, or C++.

What Success Looks Like

Success is measured through concrete outcomes:

• Higher performance on autonomous-flight benchmarks.
• Better generalization across unseen environments.
• Fewer collisions, failures, and unsafe behaviors.
• Faster and more reliable mission completion.
• Effective progress from simulation toward real flight.
• New autonomous capabilities that can be integrated into commercial projects.

In short: build increasingly capable and autonomous flying agents.

What SWARM Offers

• A competitive net salary
• 22 working days of paid vacation per year.
• 2 remote-work Fridays per month.
• Equity participation as an early member of the team.
• Relocation and administrative support for moving to Andorra.
• Generous access to compute, AI tools, and development resources.
• Professional development support based on the needs of the role.
• A transparent engineering progression framework, with regular reviews of growth and responsibilities.
• A high level of ownership and the opportunity to grow into technical leadership roles as SWARM scales.