AI Engineer - RL Post-Training and Agentic Systems Munich, Limassol

Your mission

The role

As an AI Engineer for RL Post-Training and Agentic Systems, you enter as a key member of the early AI team. You will work directly with our AI leadership and researchers to define the infrastructure we use to post-train our robot foundation models with large-scale reinforcement learning and deploy them as agentic systems across our robot fleet. You own the full loop - from distributed rollouts in simulation and on real robots, through policy optimization, to serving our models on cloud and on-prem infrastructure. The systems you build close the gap between research and deployment, turning real-world experience into better models, fast and safely. This is a full-time, permanent position.

Your profile

Key responsibilities

• Distributed RL Infrastructure: Develop scalable infrastructure for RL post-training of multi-billion-parameter robot foundation models, coordinating distributed actors, learners and reward models across GPU clusters, simulation and our robot fleet
• Model Serving: Own our cloud and on-prem serving infrastructure, delivering low-latency, high-throughput inference for both RL rollouts and production robots, including fast weight synchronization between training and inference.
• Agentic Workflows: Build the infrastructure to orchestrate, evaluate and monitor agentic systems that combine our models with planning, tool use and robot skills to complete long-horizon tasks.

Who we are looking for

We seek people who embrace the empirical nature of AI research and want to accelerate our research velocity with best-in-class tools - so that we can all make breakthroughs faster, together.

• Experience: 3+ years of hands-on experience building and operating large-scale RL or post-training systems for large models, including actor-learner architectures, asynchronous rollouts and off-policy correction.
• Core Skills: Deep expertise in RL post-training and high-performance model serving. In detail:
  • RL Post-Training: PPO, GRPO and related policy-optimization algorithms, reward modeling, and frameworks such as veRL, OpenRLHF, NeMo-RL, TRL or Ray RLlib.
  • Model Serving: Inference engines such as vLLM, SGLang or TensorRT-LLM, batching and KV-cache optimization, quantization, and serving platforms such as Ray Serve, Triton Inference Server or KServe.
  • Distributed Systems: Ray, Kubernetes and/or Slurm, fault-tolerant orchestration of heterogeneous workloads, and monitoring with Prometheus and Grafana.
• Core Stack: Python, PyTorch and/or JAX, CUDA, NCCL, and distributed training with FSDP or Megatron-Core.
• A plus: Building agentic systems and evaluation harnesses for long-horizon, multi-step tasks; robot learning in simulation and sim-to-real transfer (Isaac Sim/Isaac Lab, MuJoCo, domain randomization); edge and on-robot inference (NVIDIA Jetson/Orin/Thor, TensorRT, ONNX Runtime) under real-time latency constraints; robotics middleware and data formats (ROS 2, MCAP)
• Mindset: You balance the agility of a startup with the professional rigor required for a global industrial player. You are a curious builder who thrives in a collaborative, fast-paced research environment, and an engineer with a research mindset who understands the workflows of cutting-edge AI research and supports them with rigorous software engineering.

Why us?

• The Team: Work alongside leadership with a proven track record of building unicorn-scale companies and defining global robotics standards.
• Smart Strategy: We operate as disciplined fast followers. Capital is timed to the deployment inflection point, avoiding early-mover dead ends to scale fast on proven foundations.
• Resources: We are substantially internally funded and attract top-tier investors to avoid any capital constraints for our mission.
• Historic Opportunity: Join at the dawn of a new industry comparable to the rise of the automobile. Help shape the European champion in this trillion-dollar market.