Camera Pipeline and Image Quality Engineer

Job description

Job responsibilities

• Own the camera pipeline end to end: diagnose image quality failures, tune ISP parameters, and validate improvements across the full operating radiometric range from bright daylight to night with machine-mounted…
• Own embedded camera driver development and integration: register-level control, frame synchronization, and software interfaces that expose runtime ISP parameter control to the autonomy stack
• Characterize existing ISP pipeline behavior from first principles: identify root causes of image quality failures using parameter-level access, histogram analysis, raw vs processed frame comparison, and controlled test…
• Tune ISP and imager settings (exposure, white balance, HDR sub-frame ratios, tone mapping, noise reduction) to optimize image quality for both ML perception models and remote assistance/teleoperation
• Establish and run test protocols that confirm ISP changes improve low-light performance without adversely affecting existing daytime model performance or training data compatibility
• Work closely with perception, autonomy, and sensing engineering to translate scene and platform constraints (yaw rates, lux levels, detection ranges) into concrete ISP tuning targets
• Debug camera issues from sensor/ISP register state through to captured imagery, both in the lab and in the field
• Define and drive image quality characterization methodologies (SFR/MTF, noise, dynamic range, photon transfer curves) and track performance across hardware and ISP firmware generations

Job requirements

• Hands-on experience tuning ISP pipelines on real hardware (auto exposure, auto white balance, tone mapping, demosaicing, noise reduction, and HDR fusion) with a track record of diagnosing and correcting failure modes…
• Deep familiarity with AE algorithm internals: histogram weighting, metering zone selection, exposure ratio control in multi-exposure HDR pipelines, and lux estimation, and how these interact with scenes containing…
• Hands-on experience writing or integrating embedded camera drivers (V4L2, MIPI CSI-2, GMSL, I2C) and building the tooling and software interfaces that expose ISP and imager control to an autonomy software stack
• Familiarity with camera data pipelines on embedded platforms: frame synchronization, timestamping, compression, bandwidth management, and integration with autonomy middleware (ROS2 or similar)
• Understanding of how ISP tuning choices affect downstream ML/perception model performance, and experience validating that pipeline changes do not degrade existing model behavior
• Working knowledge of camera sensor fundamentals (CMOS architecture, shutter types, CFA patterns, dynamic range, sensitivity, and binning) sufficient to reason about how sensor choice and configuration interact with ISP…
• Working knowledge of radiometry sufficient to interpret photon budget models, SNR predictions, and motion-blur constraints as inputs to ISP tuning requirements
• Strong data analysis skills, including experience working with large datasets, building quantitative models, and using statistical methods to characterize real-world system behavior