SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Bedrock Robotics is deploying autonomous systems on heavy construction machinery across the country, backed by $350M in funding. The company is led by industry veterans from Waymo, Segment, and Uber Freight, and is focused on bringing advanced autonomy to the construction industry to address labor shortages and accelerate infrastructure projects.
You will own the full camera pipeline for autonomous construction machines, from embedded drivers and data interfaces through ISP (Image Signal Processor) tuning. The role requires ensuring cameras deliver usable imagery for both machine learning perception models and human teleoperation across challenging real-world lighting conditions—full sun to complete darkness, high dynamic range, dust, glare, and unpredictable site lighting.
Key responsibilities include:
- Diagnosing and correcting image quality failures across the full radiometric range from bright daylight to night
- Tuning ISP parameters (auto exposure, auto white balance, tone mapping, demosaicing, noise reduction, HDR fusion) on real hardware
- Developing and integrating embedded camera drivers (V4L2, MIPI CSI-2, GMSL, I2C) and building software interfaces that expose ISP control to the autonomy stack
- Characterizing ISP pipeline behavior from first principles using parameter-level access, histogram analysis, and controlled test scenes
- Establishing test protocols to validate that ISP changes improve low-light performance without degrading daytime model performance
- Collaborating with perception, autonomy, and sensing teams to translate scene constraints into concrete ISP tuning targets
- Debugging camera issues from sensor/ISP register state through captured imagery in lab and field environments
- Defining image quality characterization methodologies (SFR/MTF, noise, dynamic range, photon transfer curves) and tracking performance across hardware generations
- Managing vendor relationships with ISP, camera module, and embedded compute suppliers
Required qualifications include 8+ years of hands-on experience in ISP tuning, embedded camera systems, or image quality engineering. You must have deep familiarity with AE algorithm internals, camera sensor fundamentals (CMOS architecture, shutter types, CFA patterns), radiometry, and strong data analysis skills. Experience with embedded platforms, ROS2 or similar autonomy middleware, and C/C++/Python is essential.
