The Search & AI Platform is building the critical infrastructure that makes Bing and Copilot possible on a global scale, impacting individual's lives in scenarios such as Web Search, AI chat, AI grounding, and local deal finding. The Principal Software Engineer will collaborate in building the deep learning inferencing system using in-house and open-source technologies, and working on performance analysis, key components scaling up, distributed system debugging, and new feature/algorithm development.
Requirements
- 6+ years of experience on distributed systems (e.g., communication optimization, network architecture design, network programming).
- Experience on distributed systems (e.g., communication optimization, network architecture design, network programming).
- 8+ years of experience on high performance computing (e.g., cache/memory optimization, high-performance GPU programming, compiler-based optimization, fine-grained parallel library and runtime) or distributed systems (e.g., communication optimization, network architecture design, network programming).
- Experience on deep learning model and large language model (LLM) serving.
- Experience on performance analysis and optimization for both CPUs and GPUs, as well as good understanding on software-hardware codesign.
Responsibilities
- collaborating in building the deep learning inferencing system using in-house and open-source technologies
- working on performance analysis
- key components scaling up
- distributed system debugging
- new feature/algorithm development
- Design, and develop large scale distributed software services and solutions in collaboration with teammates
- Design services with a solid focus on high availability and scalability, robustness, and observability
Other
- 4 days / week in-office
- Travel 0-25%
- Ability to meet Microsoft, customer and/or government security screening requirements are required for this role.
- This position will be required to pass the Microsoft Cloud background check upon hire/transfer and every two years thereafter.
- Work collaboratively and communicate effectively with engineering, data scientists and product teams.
