Software Engineer, Simulation (Digital Twins)

Laelaps AI is a company focused on developing intelligent software for physical security, leveraging advancements in AI, cloud computing, and hardware. The company aims to create a safer, more secure world by enabling robots and sensors to handle critical tasks autonomously. The role of Software Engineer, Simulation (Digital Twins) is a key position within the company, tasked with building and improving simulation environments for digital twins of real-world sites. The main day-to-day responsibilities involve creating a pipeline that converts lidar maps into simulation-ready digital twins, complete with geometry, semantics, materials, and sensors. This includes developing training environments for reinforcement learning (RL) skills, generating rare event scenarios, and ensuring that sensor simulations are realistic enough to transfer learning from the digital twin to real robots. The engineer will also work on closing the gap between simulation and real-world deployments and scaling the simulation environments to handle many parallel tasks efficiently. To qualify for this role, candidates must have at least three years of experience in building robotics simulation or RL training environments. Proficiency in Isaac Sim or Isaac Lab is required, with experience in MuJoCo or Gazebo also considered. Strong Python skills are essential, and knowledge of C++ is a plus. Familiarity with 3D data, including point clouds, meshes, and scene formats like USD, is necessary. Additionally, experience with GPU-parallel simulation and reinforcement learning workflows is required. Nice-to-have skills include experience with legged robots, photogrammetry, Gaussian splatting, or other scene reconstruction techniques. Knowledge of ROS 2 and thermal or lidar sensor modeling is also beneficial. The role is based in Zurich, offering a collaborative environment with a strong founding team, including PhD-level co-founders in AI, Robotics, and Physics. The company provides a competitive equity/salary package and a culture that values both serious work and a fun atmosphere. What the role asks for: - 3+ years building robotics simulation or RL training environments. - Hands-on with Isaac Sim or Isaac Lab. - Strong Python. - 3D data: point clouds, meshes, scene formats such as USD. - GPU-parallel simulation and reinforcement learning workflows. - MuJoCo or Gazebo experience, nice to have. - C++ knowledge, nice to have. - Experience with legged robots, nice to have. - Photogrammetry, Gaussian splatting or other scene reconstruction, nice to have. - ROS 2, nice to have. - Thermal or lidar sensor modelling, nice to have.