Remote ML Perception Engineer for Autonomous Driving

Aimlroles; Sunnyvale (CA); Remote; USD 171,000 - 261,000; Full time

Job summary

General Motors is seeking a highly skilled ML researcher to design, train and evaluate perception models using multi‑modal data, including camera, lidar, and radar. You will work on object detection, segmentation, tracking, and motion prediction, integrating models into production systems with a focus on safety and robustness.

The role emphasizes collaboration with Safety, Systems Engineering, and Product teams, development of efficient training/inference pipelines, and applying advanced

Qualifications

• BS, MS, or PhD in ML, Robotics, CS, or related field, or equivalent practical experience.
• 3–5 years of experience in ML perception, prediction, or autonomous driving.
• Strong experience with multi-modal sensor data (camera, lidar, radar) and data preprocessing.

Responsibilities

• Design, train, and evaluate ML perception models for detection and segmentation.
• Develop and maintain the secondary stack perception model for fallback autonomy.
• Implement training and inference pipelines with pruning, quantization, distillation.
• Collaborate with software and infra engineers to deploy models into production systems.
• Define ML metrics and drive experiments to improve robustness.

Skills

ML perception

Multi-modal data

Python

C++

Deep learning

Object detection

Tracking

Model optimization

PyTorch

TensorFlow

JAX

ROS

Education

ML/Robotics BS/MS/PhD

Tools

PyTorch

TensorFlow

JAX

ROS

Job description

General Motors is seeking a highly skilled ML researcher to design, train and evaluate perception models using multi‑modal data, including camera, lidar, and radar. You will work on object detection, segmentation, tracking, and motion prediction, integrating models into production systems with a focus on safety and robustness.

The role emphasizes collaboration with Safety, Systems Engineering, and Product teams, development of efficient training/inference pipelines, and applying advanced