2 days ago
Zürich, SwitzerlandSenior
Responsibilities
- Design, implement, and optimize real-time onboard perception algorithms for detection, temporal association, classification, and state estimation.
- Train, retrain, and evaluate neural networks, including dataset curation, labeling strategy, and evaluation methodology.
- Optimize models and pipelines for embedded inference within latency, throughput, and compute budgets.
- Develop camera calibration and multi-camera and multi-sensor synchronization procedures.
- Fuse visual, inertial, and platform-state data using filtering and estimation techniques for robust performance under egomotion.
- Plan and conduct field tests, take algorithms from bench to flight, analyze real-world logs and video, and improve system performance.
- Build repeatable evaluation and regression methods for measuring perception accuracy across iterations.
- Advise on camera, lens, sensor, and compute selection using real performance data.
- Own computer vision deliverables end to end, align requirements across teams, and communicate with stakeholders through customer delivery.
- Mentor and coach other engineers and contribute to product quality.
Requirements
- Degree in Computer Science, Electrical Engineering, Robotics, or a related field, or equivalent practical experience.
- Hands-on experience deploying computer vision systems on resource-constrained robotic or aerial platforms.
- Strong grounding in feature detection and matching, optical flow, camera models and calibration, and multiple-view geometry.
- Experience training and deploying convolutional neural networks for detection or classification, including data collection, labeling quality, retraining, and evaluation.
- Experience debugging sensor timing, timestamps, exposure timing, synchronization, and latency issues.
- Proficiency in C++, Python, and OpenCV.
- Track record of taking computer vision work from problem definition through real-world deployment.
- Willingness to conduct field work, travel to test sites, and debug using real-world logs and video.
- Ability to lead across teams, navigate ambiguity, and collaborate with engineering, product, and external stakeholders.
- EU citizenship is required.
- Preferred experience includes state estimation, sensor fusion, Kalman and extended Kalman filters, IMU integration, visual-inertial odometry, visual SLAM, ROS2, embedded inference runtimes, quantization, pruning, GPU or NPU acceleration, perception regression testing, aerial platforms, PX4, and MAVLink-based systems.
Benefits
- Flexible working hours.
- Stock options.
- Generous holiday allowance.
- Company pension plan.
- Car parking.
- Enhanced maternity and paternity leave.
- Mental health and wellbeing support.
- Learning and development opportunities.
Categories
About Auterion
Auterion builds an operating system and software platform for autonomous robots and drones, combining onboard autonomy with cloud-based fleet management for enterprise and government operators. It monetizes through commercial subscriptions, integrations, and support for its PX4-based stack used across delivery, inspection, and defense applications. Founded in 2017 and headquartered in Arlington, Virginia, Auterion is privately held.