Image Quality Engineer, ISP Tuning

About the company

Efference provides high-performance perception and compute infrastructure for training and deploying robot models in the cloud. Our vision is to make robotic deployments scalable with shared compute.

We approach the problem by owning the entire path from the image sensor to the datacenter GPU. We shrink the load at the sensor, fix the uplink so robot traffic is fast and reliable, and clock synchronize the system so observations arrive time-aligned, ready for inference.

About the role

You'll tune Efference camera pipelines to produce consistent, useful images across the conditions robots encounter. You'll evaluate image quality in the lab and field, balancing detail, noise, exposure, and temporal behavior with the needs of compression and downstream perception.

What you’ll do

• Tune exposure and white balance, lens shading, color correction, denoising, sharpening, gamma, and tone curves across sensor modes and lighting conditions.
• Build and maintain representative scenes and capture sets, including motion, low light, mixed illumination, backlighting, and difficult materials.
• Compare tuning revisions using objective measurements and careful image review; identify texture loss, clipping, color shifts, temporal artifacts, and inconsistent cameras.
• Work with compression, VIO, depth, and evaluation engineers to test how tuning changes affect bitrate and robot performance.
• Use Python tools to analyze RAW and processed captures, run parameter sweeps, and make comparisons reproducible across firmware and calibration versions.
• Investigate unit and supplier variation, support factory calibration, and maintain versioned tuning packages with clear release criteria.

What we are looking for

• Hands-on experience with camera tuning, image-quality evaluation, or imaging projects that required systematic capture and analysis.
• Understanding of exposure, white balance, noise, sharpness, color, dynamic range, and the main blocks of an ISP.
• Ability to identify image and video artifacts precisely and design a test that separates their possible causes.
• Python or C/C++ skills and comfort working with image datasets, analysis scripts, and version control.
• Careful lab and field practice, with an ability to explain tuning decisions and tradeoffs to camera and perception engineers.

Nice to Have

• Experience with vendor ISP tuning tools, RAW/Bayer processing, or multi-camera color and exposure matching.
• Familiarity with MTF/SFR, texture, SNR, color accuracy, dynamic-range measurement, or temporal image-quality metrics.
• Experience with production camera calibration, robotics, machine vision, photography, or video capture.