北京(Beijing), Beijing, China
3 days ago
集群资源调度优化主管研究员
General Information Req # WD00066353 Career area: Research/Development Country/Region: China State: Beijing City: 北京(Beijing) Date: Wednesday, June 5, 2024 Working time: Full-time Additional Locations:  * China - Beijing - 北京(Beijing) Why Work at Lenovo We are Lenovo. We do what we say. We own what we do. We WOW our customers. 
Lenovo is a US$57 billion revenue global technology powerhouse, ranked #248 in the Fortune Global 500, and serving millions of customers every day in 180 markets. Focused on a bold vision to deliver Smarter Technology for All, Lenovo has built on its success as the world’s largest PC company with a full-stack portfolio of AI-enabled, AI-ready, and AI-optimized devices (PCs, workstations, smartphones, tablets), infrastructure (server, storage, edge, high performance computing and software defined infrastructure), software, solutions, and services. Lenovo’s continued investment in world-changing innovation is building a more equitable, trustworthy, and smarter future for everyone, everywhere. Lenovo is listed on the Hong Kong stock exchange under Lenovo Group Limited (HKSE: 992) (ADR: LNVGY). 
This transformation together with Lenovo’s world-changing innovation is building a more inclusive, trustworthy, and smarter future for everyone, everywhere. To find out more visit www.lenovo.com, and read about the latest news via our StoryHub. Description and Requirements

岗位职责:

1. 负责大模型训练资源调度,在异构集群上完成大模型的资源自动配置和自动并行

2. 设计大模型并行策略性能仿真软件,支持混合异构芯片进行大模型训练


岗位要求:

1. 全日制硕士以上学历,计算机科学与技术、人工智能等相关专业;

2. 熟练C++/Python语言、数据结构以及计算机系统结构,有AI模型性能调优经验,以及良好的工程实现能力;

3. 具备基础的GPU编程能力(CUDA / ROCm),熟悉常用的AI加速库,如NCCL/oneAPI/cudnn等;

4. 至少熟悉一种常用的深度学习框架(PyTorch/TensorFlow/Paddle/DeepSpeed等);

5. 熟悉大模型3D并行策略的原理,以及算子计算和通信开销分析手段;

6. 熟悉深度学习网络和算子底层实现细节,有模型推理或者训练调优经验.


加分项:

1. 有大模型研发和分布式训练经验

2. 熟悉Kubernetes架构以及大模型训练调度系统

3. 有大模型3D并行策略实现或者优化经验

4. 在AI或者HPC领域发表过高水平论文

Additional Locations:  * China - Beijing - 北京(Beijing) * China * China - Beijing * China - Beijing - 北京(Beijing)
Confirm your E-mail: Send Email