← Back to roles

企业微信-大模型训练框架开发工程师-AI Infra(成都/北京)

Tencent Technology (Shenzhen) Co., Ltd.

广州Posted Aug 11, 2026
Apply on Tencent Technology (Shenzhen) Co., Ltd.

Overview

Open roles
2,249
At this company
See all
Added in 30 days
528
At this company, still open

Skills named in this posting

  • C
  • C++
  • CUDA
  • PyTorch / TensorFlow
1.负责大模型训练/推理场景下的性能分析与优化,重点优化 NVIDIA GPU 及 国产芯片 的计算、显存、通信和调度瓶颈; 2.负责 CUDA / Triton / C++ 高性能算子开发与优化,包括 GEMM、Attention、MoE、KV Cache、通信融合等方向; 3.负责多卡、多机训练/推理性能调优,优化 NCCL、NVLink、NVSwitch、InfiniBand/RDMA 等通信链路; 4.负责基于 NVIDIA GPU 和 国产芯加速卡 的性能调优,提升 GPU 利用率、吞吐、延迟和显存效率; 5.负责 PyTorch、TensorRT-LLM、vLLM、SGLang、Megatron-LM 等框架下的大模型性能优化落地; 6.结合 Nsight Systems、Nsight Compute、CUDA Profiler 等工具进行性能分析,定位并解决性能瓶颈。 1.熟悉 C/C++、CUDA 或 Triton 编程,有 GPU 算子开发或性能优化经验; 2.熟悉 NVIDIA GPU 架构,理解 SM、Warp、Tensor Core、Shared Memory、Memory Coalescing、Occupancy 等概念; 3.熟悉大模型训练或推理流程,了解 Attention、GEMM、MoE、KV Cache、并行策略等核心模块; 4.熟悉分布式训练/推理优化,理解 TP、PP、DP、EP、NCCL、RDMA 等技术; 5.有实际 profiling 和性能调优经验,能够通过工具定位瓶颈并给出优化方案; 6.有 NVIDIA GPU 或 国产加速卡 实际优化经验者优先; 7.不要求候选人覆盖 AI 全栈所有方向,只要在大模型训练/推理、GPU 算子优化、分布式通信、性能调优、推理框架等任一方向有较深入理解和实践经验,均欢迎投递。

Apply on the employer's site

Original posting: careers.tencent.com

Apply on Tencent Technology (Shenzhen) Co., Ltd.

RISC-V ecosystem

Tencent Technology (Shenzhen) Co., Ltd. in the RISC-V ecosystem

RISC-V International: Premier TSC

Tencent Holdings Ltd. is a Chinese multinational technology and entertainment conglomerate and holding company headquartered in Shenzhen.

  • Tencent joined RISC-V International as China pushed for open-instruction-set alternatives (widely reported)
  • Its TencentOS Tiny IoT operating system is used on RISC-V MCUs, e.g. tutorials pairing TencentOS Tiny with the GD32V RISC-V MCU
  • Also participates in Chinese RISC-V community events such as the RISC-V Forum in Shenzhen

Same employer

View all 2,249