1 day ago
Shanghai, ChinaEntry Level
Responsibilities
- Design, develop, and optimize AI frameworks such as vLLM and PyTorch for high-performance applications.
- Implement distributed model-parallel and data-parallel algorithms and asynchronous data communication for machine-learning and deep-learning systems.
- Transform computational graph representations of neural-network models and develop machine-learning primitives in mathematical libraries.
- Profile distributed deep-learning models, identify bottlenecks, and propose performance improvements.
- Collaborate with deep-learning researchers to integrate advances into AI frameworks.
- Contribute to open-source projects while meeting industry standards.
Requirements
- Master's degree in computer science, or a bachelor's degree in computer science with at least two years of domain work experience.
- Proficiency in Python and C++.
- Foundational knowledge of vLLM, PyTorch, and deep-learning concepts.
- Demonstrated ability to debug and optimize software performance.
- Familiarity with performance-optimization technologies and techniques.
- Experience with large language models and related inference stacks is preferred.
- Background contributing to open-source projects or collaborating in open-source ecosystems is preferred.
- Strong problem-solving and complex technical-analysis skills.
- Effective written and spoken English communication skills.
- Familiarity with GPU architecture and concepts is a plus.
Benefits
- On-site work model in Shanghai, China.
- College-graduate job type.
Categories
About Intel
Intel is a public semiconductor company that designs and manufactures CPUs and other silicon for PCs, servers, networking, edge, and embedded systems, and also offers foundry services plus AI and graphics accelerators. Founded in 1968 and headquartered in Santa Clara, it sells primarily to OEMs, cloud providers, and device makers worldwide. Its portfolio includes x86 processors (Core and Xeon), Ethernet and chipset products, and the Mobileye autonomous driving platform.