TL;DR
AI Frameworks Engineer (PyTorch): Design and develop software frameworks to optimize performance across hardware backends with an accent on distributed algorithms and deep learning primitives. Focus on profiling deep learning models to identify performance bottlenecks and implementing high-efficiency primitives for neural network models.
Location: Must be based on-site in Shanghai, China
Company
A global leader in semiconductor design and manufacturing that powers the world's most transformative AI and computing solutions.
What you will do
- Design and develop AI software frameworks to optimize performance across various hardware backends.
- Implement distributed algorithms including model/data parallelism, parameter servers, and asynchronous communication.
- Transform computational graph representations and develop primitives in mathematical libraries.
- Profile distributed models to identify performance bottlenecks and propose technical solutions.
- Collaborate with researchers to integrate performance optimizations into AI frameworks.
- Contribute to open-source projects while maintaining industry standards.
Requirements
- Master's degree or higher in Computer Science, Software Engineering, or related field.
- Proficiency in C, C++, or Python programming.
- Proven experience with AI frameworks such as PyTorch.
- Solid understanding of deep learning and artificial intelligence concepts.
- Strong English language proficiency (written and spoken).
- Demonstrated ability to debug and optimize complex software systems.
Nice to have
- Experience with performance optimization technologies and techniques.
- Knowledge of large language models and related inference stacks.
- Previous contributions to open-source software ecosystems.
Culture & Benefits
- Opportunities to contribute to world-changing AI technology.
- Exposure to cutting-edge research and development environments.
- Career advancement paths within a global technology leader.
- Access to robust professional development and skill-building resources.