V&V Engineer - Multimodal AI

Quest Global; Sunnyvale (CA); On-site; USD 120,000 - 160,000; Full time

Job summary

Quest Global seeks a Systems Validation Engineer to own multi-sensor validation for next-gen wearables. You will build test capability for HW image quality feeding on-device AI, characterizing camera hardware performance and establishing repeatable setups and metrics.

Strong imaging science background and Python automation are essential. You will design validation methods across vision and sensing use cases, develop lab setups and automation, and report root causes and improvements for hardware

Qualifications

• Masters degree with 6–10 years in imaging/vision or related field.
• Hands-on validation of consumer imaging systems with objective QC metrics.
• Experience designing tests for camera and ISP outputs.

Responsibilities

• Design and build validation methods for vision and sensing use cases.
• Create lab setups, lighting, targets, fixtures and maintain calibration.
• Develop Python automation for data capture, analysis and reports.
• Propose measurement methods and pass/fail criteria when requirements are open.
• Investigate pre-production hardware issues and produce actionable reports.
• Validate multi-camera interactions under variable environmental conditions.
• Plan system-level validation for photo, video, and streaming capture per AI use cases.
• Document methods and results for repeatability and defendability.

Skills

Python programming

Optical lab skills

Image quality testing

Computer vision

Image processing

OpenCV

Scripting (bash)

Education

Masters Degree in Imaging Science/EE/CS

Tools

Lab equipment and fixtures

Automation tooling

Job description

We are seeking a Systems Validation Engineer to own multi-sensor (e.g. vision/audio) system validation for next-generation wearables. The core of the role is building test capability that does not yet exist, especially for HW image quality that feeds on-device AI features. A significant part of the job is characterizing camera hardware performance in ways that predict whether those AI features will work, then building the setups, methods and automation to measure it. Established image-quality expertise is the foundation you bring — applied through targeted studies and cross-functional optimization rather than routine metric-by-metric validation.

What you will do

• Design and build validation methods for performance across vision and other sensing AI use cases — establishing the distances, environmental conditions and scene types over which the hardware supports each feature, and characterizing where and how it fails.
• Build and own the test capability itself: specify and assemble lab setups, controlled lighting scenarios, test targets and charts, opto-mechanical fixtures and motion rigs, and keep them calibrated, documented and repeatable.
• Develop Python automation to take testing from one-off manual measurements to high-volume, repeatable runs — device control and data capture, batch gesture, audio, image and video analysis, metric extraction and automated reporting.
• Where a requirement or limit is still open, propose the measurement method, the pass/fail approach, and the data volume needed to make the result credible.
• Investigate and root-cause capture issues on pre-production hardware, and produce clear, actionable reports for hardware and software teams, with sound judgment on the conclusions and step forward.
• Validate multii-camera combined performance and interactions/coexistence under environmental and system stress, including temperature, ambient brightness extremes and thermally or power-constrained operation.
• Plan and execute system-level camera validation across photo, video and streaming capture in service of the AI use cases above; apply classic objective image-quality metrics (e.g. MTF) through targeted DOEs and cross-functional optimization support, rather than routine full-coverage validation runs.
• Document methods and results so that tests can be repeated and results defended by others.

What You Will Bring:

• Masters Degree with 6-10 years of experience and background in Imaging Science,Computer Science, Image Processing, Computer vision, Optics, Electrical Engineering, or a related field, with substantial relevant industry experience in ISP, computer vision or Image Quality Testing.
• Demonstrated hands-on experience validating consumer imaging systems against objective image quality metrics, with working knowledge of industry image quality test protocols, charts and evaluation tools.
• Practical understanding of how vision AI features consume camera output — enough to design tests that expose camera-side limitations, and to reason about accuracy, false detections, and the conditions under which a feature degrades. Model development experience is not required.
• Strong Python, with demonstrated experience building test automation and image or video analysis tooling, not only running existing scripts.
• Hands-on optical lab capability: creating, aligning, calibrating and maintaining sensitive measurement setups, working with controlled illumination, targets, opto-mechanics and motion control components.
• Comfortable working on pre-production hardware: device bring-up, flashing, shell scripting, log capture and scripted data acquisition.
• Strong analytical judgement and clear reporting. You can explain a measurement, and defend it or revisit it and drive towards clarity when your result is challenged.
• Able to work with ambiguity, incomplete specifications and shifting priorities, and to juggle competing requests from a large cross-functional team.
• Working knowledge of camera geometric calibration — interpreting and applying an intrinsic/distortion calibration file to dewarp imagery (e.g. with OpenCV), and familiarity with camera-IMU and cross-sensor temporal synchronization

Preferred qualifications

• Experience validating camera-driven perception or AI features on an embedded or wearable device.
• Understanding of system-level interactions across the imaging pipeline — sensor, optics, ISP, and the downstream consumers of image data.
• Experience building ground-truth test sets and reasoning about sample size and statistical confidence in validation results.
• Exposure to SLAM / visual-inertial (6DoF) localization, gaze/eye-tracking or hand-tracking — including the geometric-accuracy and calibration considerations these systems depend on, and test design involving human subjects and inter-subject variability.
• Experience with subjective and perceptual image quality evaluation alongside objective metrics.
• Familiarity with image sensor and optics hardware development and the associated evaluation methodologies.
• Prototyping skill with imaging test targets, custom scene setups and device interface fixtures.
• Basic familiarity with optical simulation or mechanical CAD tools for designing test rigs.