Senior ML Engineer, Motion Planning — Autonomous Driving

Didi Labs; San Jose (CA); On-site; USD 170,000 - 339,000; Full time

Job summary

DiDi Labs is seeking an experienced Machine Learning Engineer to advance the ML-powered planning system for Voyager's autonomous driving platform. You will leverage frontier ML techniques with access to petabytes of real-world driving data.

You will collaborate with ML Infrastructure, Perception, Prediction, Planning, and Systems teams to deploy robust, scalable models under onboard compute and safety constraints.

Qualifications

• BS, MS, or PhD in Computer Science, Machine Learning, Robotics, Electrical/Mechanical Engineering, or related field, or equivalent practical experience.
• 3+ years in machine learning, autonomous driving, robotics, or related fields (Senior) or 5+ years (Staff).
• Proven track record of developing and deploying ML models in production, ideally for real-time, embedded, or autonomous driving systems.
• Strong proficiency in Python and C++, with production-grade software development.
• Deep understanding of transformers, BEV/3D representations, and trajectory prediction/generation models.
• Foundations in linear algebra, control theory, or motion planning (path/trajectory generation, lane/road geometry).
• Experience applying ML to motion planning, robotics, or related decision-making problems.
• Ability to own ambiguous technical challenges end-to-end, balancing research innovation with production constraints.

Responsibilities

• Develop ML-powered systems that generate driving behavior decisions, scene understanding, and trajectories.
• Frame driving challenges as ML problems; apply imitation learning, RL, learned planners, and end-to-end models.
• Design data-driven algorithms to improve safety, generalization, and reliability of autonomous driving.
• Develop hybrid planning that integrates ML with rule-based and optimization-based planning.
• Evaluate large-scale ML models using real-world data and simulation within onboard constraints.
• Build iteration pipelines, scenario mining, closed-loop simulation, and metrics tied to real-world performance.
• Collaborate with ML Infrastructure, Perception, Prediction, Planning, Mapping, and Systems Engineering teams.

Skills

Python

C++

PyTorch

JAX

TensorFlow

Reinforcement learning

Imitation learning

Trajectory planning

Motion planning

Education

BS/MS/PhD in CS/ML/Robotics or related

Tools

PyTorch

JAX

TensorFlow

CUDA

TensorRT

Job description

DiDi Labs is seeking an experienced Machine Learning Engineer to advance the ML-powered planning system for Voyager's autonomous driving platform. You will leverage frontier ML techniques with access to petabytes of real-world driving data.

You will collaborate with ML Infrastructure, Perception, Prediction, Planning, and Systems teams to deploy robust, scalable models under onboard compute and safety constraints.