SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Helsing is a defence AI company on a mission to protect democracies through technological leadership in AI. This role sits at the intersection of AI, GPU computing, and physics-based simulation, building the core simulator that trains reinforcement-learning agents for autonomous air combat systems.
You will own the GPU-based simulator that scales to thousands of parallel environments, enabling RL agents to learn complex air-autonomy behaviors. Your responsibilities span multiple technical domains: developing high-fidelity flight dynamics and sensing models, implementing domain randomization and sim-to-real transfer techniques, and building the training infrastructure that makes large-scale RL feasible. You'll collaborate closely with AI specialists and control engineers to ensure the simulator meets both training objectives and real-world mission requirements.
Key day-to-day activities include scaling GPU simulation infrastructure, improving model fidelity and performance, developing RL training tooling, and enabling policy export to downstream systems and physical hardware. You'll own code quality and shape the shared libraries that the broader team depends on.
The ideal candidate holds an MSc or PhD in Computer Science, Machine Learning, Robotics, Control Engineering, or related field. You need solid software engineering fundamentals (Python, Rust, or modern C++), production deployment experience (testing, monitoring), and deep expertise in at least one of: GPU programming, high-performance simulation, physics/flight-dynamics modeling, or navigation and control. You care about code quality, maintainability, and the mission's real operational impact.
Nice-to-have skills include AI/ML knowledge (understanding model training and evaluation) and breadth across multiple specialist areas (e.g., combining GPU programming with simulation environment design). Helsing explicitly welcomes applications from underrepresented groups in defense tech and encourages candidates to apply even if they don't check every box.
