Senior Software Engineer – Autonomy Evaluation
Company: Gm
Location: Sunnyvale, California, United States of America (Remote)
Type: Full-time
Remote: Yes
Posted: 2026-08-31
About this role
Job Description
General Motors is a global leader in advanced driver assistance. With Super Cruise hands-free technology in more than 500,000 Super Cruise–equipped vehicles on the road and over 700 million hands-free miles driven, GM is proving that automation can be trusted, intuitive, and helpful. GM has the global reach to bring cutting-edge advances to everyday drivers at unprecedented scale. Join us to help deliver the next generation of safe and delightful personal autonomous vehicle experiences.
About the Organization
The Evaluation team builds and evolves the evaluation ecosystem that powers the development and scaling of GM’s autonomous driving technology. We develop metrics, automated workflows, and analysis approaches that enable data-driven decisions across AV development and verification. Partnering with Autonomy, Simulation, Systems, and Safety teams, we act as system-level integrators and arbiters of end-to-end AV quality.
We own large-scale test scenario libraries, continuous evaluation pipelines, and critical risk assessment and release-gating components, treating road testing, data mining, training, and metrics as first-class use cases in a unified analytics framework. By joining this team, you will help shape GM’s core evaluation platforms, turn system-level results into clear feedback for engineering and leadership, and help accelerate validated AV deployment at scale.
What You’ll Do (Responsibilities)
- Architect and implement metrics and analyses to introspect autonomous driving software performance at subsystem interfaces across the autonomy stack; partner closely with autonomy developers and systems engineers.
- Design and implement analysis algorithms that summarize, aggregate, and cluster metrics produced by simulations and on-road runs of the autonomy stack.
- Propose and develop new statistical and ML methods to quantify performance and identify patterns of system and subsystem behavior across diverse scenes and operational do...