Senior ML Engineer, Perception & Embodied AI (Real-Time)

General Motors Ventures; Mountain View (CA); On-site; Confidential; Full time

Job summary

General Motors Ventures is seeking a Senior Machine Learning Engineer for Perception within the Embodied AI organization. You will own the end-to-end pipeline for safety-critical perception models, from data curation to real-time deployment on the vehicle’s compute platform.

Responsibilities include 3D object detection and tracking, real-time map detection, multi-modal sensor fusion, and scalable training infrastructure.

Qualifications

• 5+ years in Computer Vision, Deep Learning or Perception in production.
• Hands-on with PyTorch or TensorFlow for training, experimentation and debugging.
• Experience fusing data from Camera, LiDAR, and/or Radar.
• Deploying ML models for real-time embedded systems.

Responsibilities

• Own end-to-end ML perception model lifecycle (design, training, validation, deployment).
• Develop 3D object detection and tracking for vehicles, pedestrians, cyclists.
• Real-time map detection of drivable world (lanes, boundaries, signs).
• Fuse multi-modal sensor data (Camera, LiDAR, Radar).
• Build scalable ML training infra, data mining and multi-stage training.
• Optimize models for low-latency embedded hardware and data iterations.
• Debug performance degradations and edge-case failures with data-driven analysis.
• Define and monitor robust model-level metrics for development.
• Integrate perception outputs with Safety, Systems and other teams.

Skills

Computer Vision

Deep Learning

Perception

PyTorch

TensorFlow

Sensor fusion

Real-time embedded systems

Data analysis

Education

BS/MS/PhD in Computer Science, ML, Robotics

Tools

PyTorch

TensorFlow

Job description

General Motors Ventures is seeking a Senior Machine Learning Engineer for Perception within the Embodied AI organization. You will own the end-to-end pipeline for safety-critical perception models, from data curation to real-time deployment on the vehicle’s compute platform.

Responsibilities include 3D object detection and tracking, real-time map detection, multi-modal sensor fusion, and scalable training infrastructure.