Field Robotics Engineer – Data Operations (All Genders)

About the Team

The Data Operations team owns the entire data lifecycle behind STARK's AI stack: collection, acquisition, generation, curation, and management. We run our own data-collection campaigns across Europe, evaluate new sensors and platforms, and build the internal data platform that turns raw recordings into ready-to-use datasets. Everything we produce feeds directly into the perception and autonomy systems deployed on STARK's platforms — a real data advantage is built, not bought. The team is scaling up right now: real scope, direct impact, no legacy.

Your Mission

You are the technical owner of our multi-sensor recording systems — the bridge between raw hardware and the datasets our ML teams train on. You work at the critical intersection of hardware and software: integrating EO/IR cameras, lidar, radar, and IMUs over their real interfaces, adapting the drivers and ROS2 nodes that capture them, solving time synchronization down to the microsecond, and calibrating multi-sensor rigs so every stream is spatially and temporally unified. And you don't just build in the lab: you regularly join our data-collection campaigns across Europe, deploying your systems under real conditions and making sure the data we bring home is worth training on.

Responsibilities

• Own the end-to-end integration of our sensor suites — EO/IR cameras, gimbals, lidar, radar, IMU, GNSS — from electrical interfaces and wiring to working recording pipelines

• Design and implement sensor drivers and ROS2 nodes, working directly with hardware protocols including I2C, SPI, CAN, Serial, MIPI/CSI-2, GigE Vision, and Ethernet (RTSP/UDP)

• Solve temporal synchronization across distributed sensors (PTP, PPS, hardware triggering) so all streams are perfectly aligned for fusion and labeling

• Develop and execute rigorous intrinsic and extrinsic calibration routines (camera-camera, camera-IMU) to spatially unify the sensor rigs

• Build and optimize the onboard recording stack on embedded compute (NVIDIA Jetson): hardware-accelerated video pipelines (GStreamer), efficient logging (MCAP/rosbag), and storage throughput under compute, thermal, and bandwidth constraints

• Evaluate new sensors hands-on at the bench — from probing electrical interfaces with diagnostic tools to benchmarking SDKs — and turn results into clear integration decisions

• Join field data-collection campaigns across Europe: prepare and verify sensor setups, operate the recording systems on-site, diagnose hardware and software issues in real time, and validate recordings before leaving the site

• Produce integration documentation and calibration reports so every setup is reproducible by the team

Required Skills

• Degree in Robotics, Mechatronics, Electrical Engineering, Computer Science, or comparable practical experience

• Proficiency in Python and C++ for robotics and hardware interfacing

• Experience with ROS2, including deployment on physical systems

• Hands-on experience with several hardware protocols such as I2C, SPI, CAN, Serial, or Ethernet-based sensor communication

• Experience integrating cameras or other sensors on embedded Linux platforms, ideally NVIDIA Jetson

• Solid computer vision fundamentals: camera models, distortion, projection geometry, OpenCV

• Good understanding of coordinate frames, transformations, and time synchronization in distributed systems — practical calibration experience is a strong plus

• Comfortable with hardware diagnostic tools (multimeter, oscilloscope) as well as modern software workflows (Git, CI, Docker)

• Practical and structured; willingness to travel regularly for field campaigns and to work outdoors in various conditions

Nice-To-Have

• UAV or mobile robot experience; PX4/ArduPilot and MAVLink basics

• Photogrammetry + 3D Reconstruction

• Low-level driver development or customization on embedded Linux

• GStreamer hardware-accelerated pipelines, JetPack/L4T, CUDA/TensorRT

• MIPI/CSI-2 camera bring-up, ISP tuning, or Linux kernel/device tree exposure

• Calibration tooling such as Kalibr; lidar and point cloud processing (PCL, Open3D)

• Experience with detection/tracking models and modern CV pipelines (PyTorch, ONNX)

• Experience with simulation (Gazebo, Isaac Sim) or synthetic data generation