Aurora Innovation
Staff Technical Program Manager, Release Management
$163K–$290K
Mountain View, CA +2 · Hybrid · Staff · Full-time · Posted Sep 15, 2026
- Location
- Mountain View, CA
San Francisco, CA
Pittsburgh, PA - Experience
- 7+ years.
About the job
Owns software release operations and automation, coordinating release readiness, branching, triage, and safety-focused release gates as Aurora’s driverless fleet scales.
What you'll do
- Execute release trains, daily build/release triage, branch cut-offs, and hotfix deployments across software and vehicle platforms.
- Coordinate release readiness and communication across engineering, safety, hardware, and test operations teams.
- Automate release workflows and develop auditable, data-driven release gates that support safety and compliance.
What they ask for
- 7+ years of hands-on release experience in technical program management, release engineering, or systems engineering, including software-hardware integrations.
- Experience in a safety-regulated domain, with configuration management, baseline, and change-control practices.
- Hands-on experience with CI/CD build automation and monorepo build systems, such as Buildkite, GitHub Actions, and Bazel.
- Experience managing concurrent release branches and daily triage; cross-functional collaboration, occasional evening or weekend coverage, and ability to travel as business needs arise.
Nice to have
- Hands-on experience in autonomous vehicles, robotics, machine learning systems, or complex cyber-physical environments.
- Practical experience applying ML/AI tooling to release operations, such as test-failure classification or pipeline anomaly detection.
Pay
$181,000 - $290,000 per year (base salary; San Francisco, CA and Mountain View, CA); $163,000 - $261,000 per year (base salary; Pittsburgh, PA)
Tools
ISO 26262, DO-178C, ISO 13485, Buildkite, GitHub Actions, Bazel, AWS
AI focus
Use AI/ML tools for triage classification and anomaly detection, including automated test-failure classification and pipeline anomaly detection.
Model-written summary · the original posting is authoritative