Senior Machine Learning Engineer – Robotics/Swarming (All Genders)

Description

The company builds autonomous defence platforms and AI-driven software for NATO Allies and their partners, designed to run in demanding environments.

The team works fast and ships working software for constrained settings such as low-bandwidth networks, air-gapped devices and tight decision loops. Everything built goes to operators in the field.

The role sits at the core of the behaviour logic for next-generation autonomous drone swarms. The focus is on decision-making frameworks, multi-agent reinforcement learning and high-fidelity simulation rather than computer vision, and the job links machine learning models to physical flight controls so swarms can reason and coordinate in real time.

This is a hands-on individual contributor role.

Responsibilities

• Design, train and deploy decision-making frameworks using RL, imitation learning and behaviour trees for coordinated flight across fixed-wing, tube-launched and quadcopter platforms.
• Build algorithms for decentralised task allocation and multi-vehicle coordination when communication is limited or GPS is unavailable, extending the existing TDOA/RSSI localisation and mesh networking work.
• Use ROS2 SITL environments heavily to stress-test behaviour logic, neural networks and reactive behaviours before hardware, extending the current simulation work on containerised comms and leader-follower scaling.
• Build pipelines that move trained models from the GPU cluster to edge robotics hardware without losing performance, feeding the hardware phase: mesh networking on real drones and end-to-end flight tests.
• Work with the perception and flight control teams so AI behaviours connect cleanly to safety-critical C++ flight software.
• Profile and debug behaviour systems under embedded limits, keeping them stable and robust in the field on all three platform types.
• Take part in architecture discussions on autonomous decision-making, heuristic planning and multi-agent reliability.

Requirements

• A Master's or PhD in robotics, computer science, aerospace engineering or a related field, with a focus on autonomous decision-making.
• 3+ years of professional or advanced research experience in robotics AI, multi-agent reinforcement learning or autonomous behaviour modelling.
• Strong Python and C++ skills for embedded and robotics work, and comfort working near safety-critical flight code.
• Expert use of SITL workflows to validate neural networks and decision logic under variable, adversarial or degraded communications.
• Solid theoretical and practical grounding in MDPs, game theory, heuristics and trajectory or motion planning, applied to strike-capable UAV coordination.
• A proven record of taking ML models from simulation to physical edge robotics, ideally on multi-vehicle or swarm systems rather than single agents.
• Strong debugging skills in real-time, resource-constrained environments.
• Able to communicate well across autonomy, hardware and flight-software disciplines.
• Willing to travel occasionally for field testing and deployment.
• Fluent English, C1 preferred. German is an advantage but not required.
• Send your CV in English.

Conditions

Eligibility to obtain and maintain the security clearance required for the role is mandatory. Details are shared during recruitment.

Travel is occasional, for field testing and deployment.

Benefits

• Your work protects personnel and supports European and national sovereignty.
• You will work with leading experts in autonomous systems, advanced propulsion and secure cyber platforms.
• The environment is collaborative and quick-moving, and bold ideas get real R&D backing.