Engineering Manager - Autonomy Subsystem Evaluation
About The Position
The Manager of Subsystem Evaluation will lead a team of engineers and data scientists and own the team’s roadmap, execution, and staffing. The team develops evaluation metrics for autonomy subsystems—including planning, perception, prediction, and localization—that measure how subsystem errors affect driving rather than relying on accuracy alone. It also traces driving failures to the subsystem or interaction that caused them and assesses the effects of subsystem errors on driving outcomes. The role works with autonomy teams that use the metrics, the Driver Performance and Scenario Classification teams, and principal and staff engineers who lead evaluation methodology. The manager reports to the Senior Manager of Eval Applications. The position is categorized as remote/hybrid: it is remote, but employees living within 50 miles of specified GM locations are expected to report to a location at least three times per week. Travel is expected to be less than 25%. Relocation benefits may be available to eligible candidates. The role includes incentive-pay potential and eligibility for a company vehicle evaluation program after a motor vehicle report review; participants generally must purchase or lease a qualifying GM vehicle every four years, subject to limited exceptions.
Requirements
- At least 8 years of experience in autonomous systems, robotics, ML evaluation, or applied data science, including hands-on evaluation of planning, perception, prediction, or localization.
- At least 3 years managing engineers or data scientists, including hiring and performance management.
- Deep knowledge of autonomy metrics and their limitations, such as planner quality measures, detection and tracking metrics, and trajectory prediction metrics.
- Strong statistics for evaluation, including sampling, confidence intervals, experimental design, and hypothesis testing on time-series data.
- A disciplined approach to investigating anomalous data and unexpected system behavior.
- Production Python and SQL experience in team environments, and comfort reviewing designs and code.
- Ability to read C++ autonomy code to understand subsystem behavior.
- Ability to build trusted relationships with partner teams and influence them without authority using trusted data.
- Ability to turn ambiguous questions and uncertain results into clear plans and recommendations for engineering and leadership.
- Bachelor’s, master’s, or PhD in Computer Science, Robotics, Mechanical or Aerospace Engineering, Machine Learning, Statistics, Data Science, or a related field, or equivalent practical experience.
Nice To Haves
- Experience in autonomous driving or driver assistance, working with simulation and on-road data, including camera, lidar, and radar.
- Understanding of vehicle dynamics and planning and controls system design.
- Experience with causal or counterfactual analysis.
- Experience with ML interpretability, such as attribution, ablations, or embedding analysis.
- Strong data-visualization instincts and ability to make high-dimensional metrics clear to technical and non-technical audiences.
- Experience leading mixed teams of software engineers and data scientists.
Responsibilities
- Set the subsystem evaluation roadmap and quarterly plans in line with AV program milestones and release needs, prioritizing work and deciding what to stop.
- Develop metrics that weight subsystem errors by their effects on downstream decisions and driving, rather than accuracy alone.
- Trace poor driving in simulation and road data to the responsible subsystem or interaction, and measure how subsystem errors contribute to driving outcomes.
- Provide autonomy teams and release reviews with timely, trustworthy assessments of whether software or model changes helped or harmed driving, including regression detection and version-to-version comparisons.
- Validate metrics against driving outcomes and human judgment, communicate their confidence and limitations, and retire metrics that do not hold up.
- Apply VLMs and LLMs where applicable in development, triaging, and related processes.
- Present clear findings to autonomy teams and leadership, including what changed, confidence in the results, and what to fix first.
- Hire, develop, and retain engineers and data scientists through clear expectations, regular feedback, and fair performance management.
- Establish clear ownership and operating rhythms, and maintain technical quality through design reviews, code review, testing, and statistically sound analysis.
- Balance near-term program requests with longer-term investments in metrics and methods.
- Build trust with autonomy teams, align on the boundaries between component metrics and subsystem evaluation, and resolve disagreements about priorities or conflicting results.
Benefits
- Incentive pay program with payouts based on company performance, job level, and individual performance.
- Eligibility for a company vehicle evaluation program, subject to successful motor vehicle report review and program terms.
- Relocation benefits for candidates who qualify under company policy.
- Benefits supporting employee well-being at work and at home; specific benefits are described in GM Total Rewards resources.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Manager
