Multimodal Sensing and Fusion Engineer

As Multimodal Sensing and Fusion Engineer, you will own the sensing stack end to end. You will architect the multimodal perception front-end that combines electro-optical and long-wave infrared cameras, radar, LiDAR, and inertial sensing into a single, coherent input to Helsing's AI. The systems you design go into operational environments, and the quality of the data you deliver directly affects the decisions Helsing's AI supports in the field.

What you'll do

• Define the multimodal sensing architecture: select and integrate EO and LWIR cameras, radar, LiDAR, and IMUs, aligning requirements with the AI team to optimise algorithm performance and balance trade-offs across size, weight, power, cost, and robustness.
• Own sensor synchronisation: design and implement time synchronisation across heterogeneous sensors using PTP/gPTP, PPS, and hardware triggers such as Hsync and Vsync, so all modalities share a common, precise time base.
• Own sensor calibration: develop intrinsic, extrinsic, spatial, and temporal calibration procedures, including online and continuous recalibration, across cameras, radar, and IMU.
• Define safety concepts: derive failure modes, fault detection, isolation, and recovery concepts for the sensing subsystem, drawing on both defence mission-safety thinking and automotive functional-safety practice (ISO 26262, SOTIF).
• Drive system design and architecture: analyse the interplay of hardware, software, and algorithm architectures, and define the interfaces and data pipelines from sensor to AI.
• Bring hardware to life: spend hands-on time in the lab bringing up prototypes, testing performance under varying conditions, and supporting integration into the software stack.

What you'll need

• A Ph.D. or M.Sc. in Electrical Engineering, Computer Engineering, Applied Physics, Robotics, or a related field, or equivalent professional experience.
• A professional track record building multimodal sensing or sensor-fusion systems combining at least two of: cameras (EO and/or LWIR), radar, LiDAR, and IMU.
• Demonstrated understanding of system-level synchronisation, including PTP/gPTP, PPS, and hardware triggers such as Hsync and Vsync.
• Mastery of sensor calibration, intrinsic and extrinsic, spatial and temporal, including strategies for maintaining calibration across a multi-sensor rig over its operational lifetime.
• The ability to derive safety concepts for a sensing or perception subsystem, from a defence or mission-critical background, an automotive functional-safety background (ISO 26262, SOTIF), or both.
• Strong systems design instincts, comfortable reasoning about the interplay of hardware, software, and algorithms.
• Proficiency in C/C++ or Rust for low-level firmware development and hardware-software interface debugging, plus Python.

Nice to have

• Depth in one or more of the following sensor modalities: thermal imaging (LWIR design, calibration, and programming), FMCW radar signal processing (detection, clustering, and tracking), or LiDAR and point-cloud processing including cross-calibration with cameras and radar.
• IMU and inertial navigation expertise, including GNSS/INS integration, bias and error modelling, and sensor alignment.
• Sensor-fusion estimation experience: Kalman/EKF/UKF, particle filters, factor graphs, multi-object tracking, and data association.
• Low-level hardware experience: FPGA development using Verilog, SystemVerilog, or VHDL, Linux kernel driver development, or experience with high-bandwidth sensor interfaces such as MIPI CSI-2 or LVDS.
• A background in automotive, robotics, or aerospace sensor fusion, covering ADAS, AUTOSAR, ROS, or ISO 26262 and SOTIF, and familiarity with how sensor data feeds downstream computer-vision or AI pipelines.