高性能计算工程师
Moonshot AI · 上海市
Apply on Moonshot AI's siteTailor my resume
Posted
Sep 15, 2025Last verified
23 min agoEmployment
Full-timeLevel
MidFunction
Software EngineeringCUDATritonC++LLMsGPU
Open 14 months. Roles open this long are often filled, or kept open all year. Check the date on the employer's site before you apply.
岗位职责
- 负责大规模深度学习训练/推理框架中 CUDA 内核的性能优化与定制开发
- 开展与 GPU 体系结构相关的前沿探索性工作,包括新指令集、算子融合、异构计算和硬件趋势研究
- 与算法研究员合作,推动模型结构与底层实现的协同优化
任职要求
- 计算机、电子工程或相关专业本科及以上学历,或3年以上相关工作经验
- 精通 CUDA 编程模型,熟悉 Tensor Core、WGMMA、Shared Memory 等底层细节
- 具备优秀的代码实现能力(C++/CUDA/Triton/CUTLASS),良好的工程化与调试能力
- 具备良好的沟通与协作能力,能够与算法研究员、硬件工程师及跨团队成员高效配合
- 熟悉并行算法设计与优化,有大规模矩阵乘 (GEMM)、注意力 (FlashAttention/MLA)、MoE、稀疏算子优化经验者优先
- 熟悉分布式训练/推理框架(Megatron、vLLM、SGLang、TensorRT-LLM 等)者优先
- 参与过开源社区(NCCL、CUTLASS、Triton、TileLang、Megatron-LM、vLLM 等)贡献者优先
- 有 HPC 或超大规模 AI 模型训练经验者优先
Attention Head lists this role as published on Moonshot AI's own job board. Applications go to the employer.
Similar roles at other employers
- Vision Lab Engineer, East ChinaOverview Corporation · China; Shanghai, ChinaNov 18, 2026
- Agentic Engineer I - Campaigns & ContactsWATI.io · Shenzhen, Guangdong Province, ChinaNov 17, 2026
- 软件工程师(IoT智能物联方向)_ RBLCBosch Group · Shanghai, Shanghai, ChinaNov 16, 2026
- Software Engineering – AI AgentAgilent Technologies, Inc. · China-ShanghaiNov 13, 2026
- Mobile UI DesignerBjak · ChinaOct 16, 2026
All mid-level Software Engineering roles in China →
More roles at Moonshot AI
- Nov 20, 2026
- Nov 4, 2026
- Nov 2, 2026
- Oct 20, 2026
- Sep 18, 2026
- Sep 18, 2026
- Sep 18, 2026
- Sep 4, 2026