About Us

STARK is a new kind of defence technology company revolutionizing the way autonomous systems are deployed across multiple domains. We design, develop and manufacture high-performance unmanned systems that are software-defined, mass-scalable, and cost-effective. This provides our operators with a decisive edge in highly contested environments.

We're focused on delivering deployable, high-performance systems - not future promises. In a time of rising threats, STARK is bolstering the technological edge of NATO Allies and their Partners to deter aggression and defend Europe - today.

About the team

We move fast, ship real software, and operate under constraints most engineers never encounter — low-bandwidth networks, air-gapped devices, high-stakes decision loops. There is no room for abstraction for its own sake. Everything we build ends up in the hands of real operators in the field.

Our Team is dedicated to a mission of pure strikes.

Your mission

As an Senior AI Systems Engineer with a focus on Robotics and Swarming, you will play a critical role in defining the tactical brain and behavioral logic onboard next-generation autonomous drone swarms. Rather than focusing on computer vision, you will work directly with advanced behavioral frameworks, multi-agent reinforcement learning, and high-fidelity simulation environments to build robust, scalable decision-making functionality.

You will contribute as a highly skilled individual contributor—hands-on with the system—bridging the gap between machine learning models and physical flight controls, ensuring swarms can dynamically reason and coordinate in real-time. Your work will be essential to ensuring that our autonomous systems operate reliably in real-world, unpredictable environments.

Responsibilities

• Design, train, and deploy decision-making frameworks using RL, imitation learning, and behavior-tree architectures for coordinated behavior across our fixed-wing, tube-launched, and quadcopter platforms.
• Develop and optimize algorithms for decentralized task allocation, collective intelligence, and multi-vehicle strategic coordination under communication-constrained or GPS-denied conditions — building on our existing TDOA/RSSI localization and mesh networking work.
• Build and heavily utilize ROS2 SITL environments to stress-test behavioral logic, neural networks, and reactive behaviors before hardware deployment, extending our current simulation-phase epic (containerized comms, leader-follower scaling).
• Engineer pipelines to move trained models and policies off the GPU cluster and onto edge robotics hardware without performance degradation, feeding directly into our hardware-phase epic (mesh networking with real drones, end-to-end flight test).
• Collaborate closely with the perception and flight control teams to ensure AI-driven behaviors interface cleanly with safety-critical C++ flight software.
• Profile and debug behavioral system performance under embedded constraints, ensuring stability and robustness in field deployments across all three platform types.
• Contribute to system-level architecture discussions on autonomous decision-making, heuristic planning, and multi-agent reliability.

Qualifications

• Master's or Ph.D. in Robotics, Computer Science, Aerospace Engineering, or related field with emphasis on autonomous decision-making.
• 3+ years professional or advanced research experience in Robotics AI, multi-agent reinforcement learning, or autonomous behavioral modeling.
• Strong programming proficiency in Python and C++ for embedded and robotics development; comfort working alongside safety-critical flight code.
• Mastery of SITL workflows to validate neural networks and decision-making logic under variable, adversarial, or degraded-comms conditions.
• Deep theoretical and practical knowledge of MDPs, game theory, heuristics, and trajectory/motion planning, applicable to strike-capable UAV coordination.
• Proven track record moving ML models from simulation to physical edge-robotics systems — ideally on multi-vehicle or swarm platforms rather than single-agent robotics.
• Strong debugging skills in real-time, resource-constrained environments.
• Effective communicator able to work across autonomy, hardware, and flight-software disciplines.
• Willingness to travel occasionally for field testing and deployment.

For further information please reach out to Sally Grütte-Pad, Interim Lead TA Partner via [email protected]

Nice to have