Senior AI Infrastructure Engineer, Inference & Optimization

Didi Labs; San Jose (CA); On-site; USD 170,000 - 351,000; Full time

Job summary

DiDi Autonomous Driving is seeking a Senior/Sr. Staff AI Infrastructure Engineer to lead AI model deployment, optimization, and resource scheduling across on-vehicle and cloud infrastructure.

You will design high-efficiency inference pipelines and robust, low-latency systems for embedded constraints. The role requires a Master’s degree and 3–8+ years in high-performance AI infra, with expertise in C++, Python, CUDA, and modern inference engines.

Qualifications

• Master’s or higher in CS/Software/Systems Engineering or related field.
• 3–8+ years in high-performance AI infrastructure, model optimization, or embedded deployment.
• Strong proficiency in C++ and Python, CUDA/OpenMP, and low-level profiling tools.

Responsibilities

• Own deployment, optimization, and resource scheduling of vehicle-side AI models within embedded constraints.
• Lead vehicle-side system stability initiatives; diagnose and resolve complex bottlenecks and runtime anomalies.
• Architect and scale deployment environments for LLMs to support offline simulation, annotation, and validation.
• Track and evaluate cutting-edge methodologies; incorporate optimization toolchains and quantization techniques.
• Establish system-level profiling/telemetry using CUDA tools to maximize GPU utilization.
• Collaborate with AD Perception/Prediction, Cloud Infra, and Safety teams for rapid iteration.

Skills

C++

Python

CUDA

OpenMP

Profiling

System design

Root-cause analysis

GPU architectures

Education

Master's degree in CS/Engineering

Tools

TensorRT

ONNX Runtime

vLLM

SGLang

TensorRT-LLM

PyTorch

CUDA tools

Job description

DiDi Autonomous Driving is seeking a Senior/Sr. Staff AI Infrastructure Engineer to lead AI model deployment, optimization, and resource scheduling across on-vehicle and cloud infrastructure.

You will design high-efficiency inference pipelines and robust, low-latency systems for embedded constraints. The role requires a Master’s degree and 3–8+ years in high-performance AI infra, with expertise in C++, Python, CUDA, and modern inference engines.