Machine Learning Engineer - Agentic AI

DiDi•San Jose, CA

15h

About The Position

DiDi Autonomous Driving is developing Level 4 autonomous-driving technology to make transportation safer and more efficient. This role explores using AI agents to improve autonomous-driving software. Agents will propose changes to policy code, data, training recipes, and model code; evaluate results; diagnose failures; and turn verified fixes into reusable skills and data. Using DiDi Voyager's real-world driving data and large-scale evaluation infrastructure, the role aims to build an improvement loop that generalizes across scenarios, passes safety gates, and reaches vehicles. It combines machine learning, agentic systems, and autonomous driving, and contributes to both onboard architecture and the development of driving software. The position also works with prediction, planning, simulation, and onboard teams to move prototypes into production workflows.

Requirements

Strong background in machine learning, such as deep learning, reinforcement learning, or decision-making systems.
Hands-on experience building LLM agent systems, such as tool use, multi-agent orchestration, or automated evaluation loops.
Experience training or fine-tuning LLMs for agentic workflows, including trajectory data curation, supervised fine-tuning, or reinforcement learning from execution feedback.
Strong Python skills and working knowledge of C++.
Rigorous experimental habits, including controlled comparisons, held-out evaluation, and distinguishing measured results from assumptions.
Ability to own a system from research prototype through production.

Nice To Haves

Experience building an agentic harness or self-improving loop for a specific product or domain.
Experience in autonomous driving or robotics, especially planning, prediction, simulation, or sim-to-real transfer.
Familiarity with code-as-policy, program synthesis, LLM-driven optimization, or evolutionary search.
Publications at top machine learning or robotics venues, such as NeurIPS, ICML, ICLR, CVPR, RSS, or CoRL.
A BS, MS, or PhD in Computer Science, Machine Learning, Robotics, or a related field.

Responsibilities

Build agentic harnesses that provide agents permissioned access to the driving stack, data, training infrastructure, and evaluation environments, with automatic scenario setup and pass/fail checks.
Develop recursive self-improvement loops involving failure mining, candidate generation, evolutionary search, and targeted data generation to address weaknesses in the driving policy.
Curate a reusable skill library, ensuring verified fixes help before they are promoted for future tasks.
Create training data from development trajectories and fine-tune LLMs to work effectively with the software stack, tools, and codebase.
Advance agent-as-policy approaches by using agents to write policy logic, curate data, adjust training recipes and objectives, and trace driving failures to their causes.
Shape the onboard architecture by determining what runs on the vehicle versus offline within constraints on latency, compute, determinism, and safety.
Collaborate with prediction, planning, simulation, and onboard teams to turn prototypes into production workflows.