Senior Autonomy Evaluation Engineer

General Motors LLC; Sunnyvale (CA); Hybrid; USD 145,000 - 221,000; Full time

Benefits offered by this job

Hybrid work model

Relocation benefits

Health & retirement benefits

Job summary

General Motors, a global leader in autonomous driving technology, seeks a seasoned analytics engineer to build evaluation metrics and dashboards. You will collaborate with autonomy, simulation, systems, and safety teams to enable data-driven decisions and scalable validation of GM’s AV stack.

You will implement analytics, experiments, and visualization tools, while mentoring peers and driving architecture decisions for end-to-end evaluation pipelines.

Qualifications

• 5+ years of robotics or autonomous systems software experience.
• 3+ years evaluating dynamic systems using numerical/ML approaches.
• Strong Python production development experience (testing, performance, review).
• Proficiency with Pandas, NumPy, SciPy and visualization for large-scale data.
• Comfort with reading/debugging C++ codebases and instrumenting algorithms.

Responsibilities

• Architect and implement metrics and analyses across the autonomy stack; partner with developers and systems engineers.
• Design analysis algorithms to summarize, aggregate, and cluster metrics from simulations and on-road runs.
• Propose ML/statistical methods to quantify performance and identify patterns across scenes.

Skills

Robotics experience

Dynamic systems analysis

Python programming

Data analysis with Python

C++ programming

Anomaly detection

Technical leadership

Education

Degree in CS/Robotics/ML

Tools

ROS

SQL

PyTorch

C++

Job description

General Motors, a global leader in autonomous driving technology, seeks a seasoned analytics engineer to build evaluation metrics and dashboards. You will collaborate with autonomy, simulation, systems, and safety teams to enable data-driven decisions and scalable validation of GM’s AV stack.

You will implement analytics, experiments, and visualization tools, while mentoring peers and driving architecture decisions for end-to-end evaluation pipelines.