AI Infrastructure Architect for GPU-Accelerated HPC

Jingdong Development France; United Kingdom; On-site; GBP 120,000 - 180,000; Full time

Job summary

JD.com is seeking a senior AI infrastructure engineer to design and drive architecture for cross-border e-commerce platforms. You will collaborate across teams to ensure scalable, reliable AI workloads and high-performance computing foundations.

The role emphasizes GPU/NPU performance, fault tolerance, and observability, with hands-on work on CUDA/C++, Python, PyTorch ecosystems, and modern scheduling systems like Kubernetes or Ray.

Qualifications

• 3+ years in AI infrastructure, distributed systems, HPC, or large-scale cloud platforms.
• Strong Python proficiency with solid data structures and algorithms.
• Hands-on CUDA/C/C++ or Triton experience for GPU/NPU architectures.
• Bachelor's degree or higher in CS/EE/Applied Mathematics or related field.
• Familiarity with PyTorch, Megatron-LM, or DeepSpeed is preferred.

Responsibilities

• Lead top-level architectural design for complex AI infra projects.
• Optimize GPU resource scheduling within Kubernetes or Ray.
• Resolve GPU cluster fault tolerance and elastic checkpointing.
• Develop cluster efficiency models and monitor resource usage.

Skills

AI infrastructure

Distributed systems

HPC

Cloud platform development

Python

Algorithms & data structures

CUDA/C++/Triton

PyTorch Megatron-LM DeepSpeed

Education

Bachelor's degree in CS/EE/Applied Mathematics or related field

Tools

CUDA

PyTorch

Megatron-LM

DeepSpeed

Kubernetes

Ray

Job description

JD.com is seeking a senior AI infrastructure engineer to design and drive architecture for cross-border e-commerce platforms. You will collaborate across teams to ensure scalable, reliable AI workloads and high-performance computing foundations.

The role emphasizes GPU/NPU performance, fault tolerance, and observability, with hands-on work on CUDA/C++, Python, PyTorch ecosystems, and modern scheduling systems like Kubernetes or Ray.