AI Software Development Eng.

AMD
Beijing
Workplace: OnsiteFull timeCNY 581,350 - 830,500 annuallyFunction: Data Science & Machine LearningEducation: bachelorsSkills: ["Collaboration","Problem-solving","Independent execution","Analytical thinking","Software engineering best practices"]

Optimize and develop deep learning frameworks for AMD GPUs, improving GPU kernel performance and enabling efficient training and inference at scale across multi-GPU and multi-node systems. Collaborate with internal GPU software, compiler, and math library teams, while integrating upstream open-source compiler technologies. Contribute to SGLang development, tune large-scale deep learning models, and integrate graph compilers such as XLA and TorchDynamo to align runtime execution with AMD hardware performance goals.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
AMD
AMD
10 hours ago

AI Software Development Eng.

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 8 hours agoStatus: Live

Job Summary

Optimize and develop deep learning frameworks for AMD GPUs, improving GPU kernel performance and enabling efficient training and inference at scale across multi-GPU and multi-node systems. Collaborate with internal GPU software, compiler, and math library teams, while integrating upstream open-source compiler technologies. Contribute to SGLang development, tune large-scale deep learning models, and integrate graph compilers such as XLA and TorchDynamo to align runtime execution with AMD hardware performance goals.
Location: Beijing
Workplace: Onsite
Employment Type: Full time
Job Function: Data Science & Machine Learning

Key Responsibilities

  • •Enhance performance of deep learning frameworks (e.g., TensorFlow and PyTorch) on AMD GPUs through upstream open-source contributions.
  • •Develop and optimize large-scale training and inference models for optimal performance on AMD hardware.
  • •Design, implement, and optimize high-performance GPU kernels for AI operator efficiency using HIP, Triton, or relevant tools.
  • •Work with internal compiler and GPU library teams to integrate kernel-level optimizations with end-to-end performance goals.
  • •Contribute to SGLang development and optimize distributed execution across multi-GPU and multi-node systems, including inference parallelism and graph compiler integration (e.g., XLA, TorchDynamo).

Pay and Benefits

Salary: CNY 581,350 - 830,500 annually

Key Requirements

  • •Expert in C++ and/or Python with the ability to debug, profile, and optimize performance-critical code in Linux environments.
  • •Strong hands-on experience optimizing deep learning frameworks such as PyTorch or TensorFlow on AMD GPUs.
  • •Knowledge of GPU/kernel development using HIP, Triton, or related GPU programming tools to improve AI operator efficiency.
  • •Familiarity with compiler and GPU architecture concepts, including LLVM, MLIR, or ROCm (preferred).
  • •Experience scaling heterogeneous workloads (CPU+GPU) using distributed training or inference strategies and collective communication approaches (preferred).
Experience:Deep learningGPUsOpen source
Education:Bachelor's in Computer Science, Computer Engineering, Electrical Engineering, or a related field
Skills:CollaborationProblem-solvingIndependent executionAnalytical thinkingSoftware engineering best practices
Languages:En-us
Tech Stack:C++PythonLinuxTensorFlowPyTorchSGLangHIPTritonLLVMMLIRROCmCUDAXLATorchDynamoGraph compilersGCNCDNA

Company Brief

AMD
Designs and produces semiconductor products including CPUs, GPUs, and adaptive SoCs for consumer, enterprise, and embedded markets, competing across PCs, data centers, and gaming industries.
Industry: Electronics Manufacturing
Company Size: Enterprise (1,001+ employees)
Revenue: USD 1B+
Growth: Public Company
Valuation: Public Company (Market Cap in USD)
Funding: IPO / Publicly Listed
Headquarters: Santa Clara, United States
Founded: 1969
Glassdoor
Glassdoor: 3.9
WebsiteLinkedIn