Software Engineer, Robot Autonomy (Localisation & State Estimation)

Laelaps AI is a Zurich‑based startup developing intelligent software that powers autonomous security robots and sensors. The company's mission is to let machines handle dangerous, critical tasks, creating safer and more resilient environments. As a Software Engineer focused on Robot Autonomy, you will own the state‑estimation stack that tells each robot where it is and how it moves, fusing data from IMUs, cameras, LiDAR, GNSS and platform‑specific odometry. Your daily work will involve designing and delivering real‑time estimators for position, orientation, velocity and angular velocity across multiple robot platforms. You will integrate heterogeneous sensor streams, manage noise, asynchronous timing, calibration errors and occasional drop‑outs, and build reusable components that adapt to different dynamics. The role also requires improving localisation in GNSS‑denied or otherwise degraded conditions, linking the estimator to perception, planning and control modules, and providing reliable uncertainty information. You will investigate failures using logs, datasets, simulation and field tests, develop repeatable evaluation workflows, and collaborate closely with the broader Robot Autonomy team and forward‑deployed engineers to turn operational feedback into system upgrades. The position demands a Master's degree or PhD in a relevant engineering discipline, or equivalent hands‑on experience, plus at least three years of proven work building or deploying state‑estimation, localisation or odometry systems for robots or autonomous platforms. You must have a solid grasp of nonlinear estimation, sensor fusion, Kalman filtering and uncertainty representation, and be comfortable handling real‑world sensor issues such as noise, calibration, synchronization and data loss. Strong C++ and Python programming skills are required, together with experience developing robotics software in ROS 2. You should be adept at analysing estimator failures, debugging across sensors and software, and evaluating robotics systems using recorded data, simulation or physical robots. Robust software‑engineering habits, including version control, reproducible experiments and disciplined performance testing, are also essential. Experience with both ground and aerial robots, visual‑inertial or LiDAR‑inertial odometry, SLAM, GNSS/INS integration, or optimisation‑based tools like GTSAM or Ceres is considered a plus. Familiarity with robot dynamics, operation in challenging environments (low light, dust, rain, high vibration) and field monitoring of deployed systems are also valued. Publications at top robotics venues are welcomed but not required. The role offers end‑to‑end product ownership, a mission‑driven focus on autonomous security, a clear growth path within a ground‑floor team, and direct interaction with PhD‑level founders. Laelaps AI provides a competitive equity‑and‑salary package and an international, fun‑yet‑serious culture. What the role asks for: - Master's degree, PhD or equivalent practical experience - 3+ years building state‑estimation or localisation systems - Strong C++ programming skills - Strong Python programming skills - Experience with ROS 2 - Knowledge of nonlinear state estimation and Kalman filtering - Ability to handle real sensor noise, calibration and drop‑outs - Proven debugging of estimation failures - Experience evaluating with recorded data, simulation or robots - Strong software‑engineering practices (version control, reproducible experiments) - Nice to have: ground and aerial robot experience - Nice to have: visual‑inertial or LiDAR‑inertial odometry expertise - Nice to have: familiarity with GTSAM or Ceres - Nice to have: work in GNSS‑denied or harsh environmental conditions - Nice to have: field monitoring of deployed robotics systems - Nice to have: publications at top robotics venues