微软Principal Software Engineering Manager- GPU Inference Optimization
任职要求
• Bachelor's degree in computer science or related technical field AND 5+ years technical engineering experience with coding in languages including, but not limited to, C/C++, CUDA, ROCm or equivalent experience • Practical Experience writing new GPU kernels, going beyond experience of GPU workloads with existing library kernels • Quick learning, good communication (fluent in English) and solid problem-solving skills • Cross-team collaboration skills and the desire to collaborate in a team of researchers and developers • Experience in low-level performance analysis and optimization, including proficiency using GPU profiling tools such as NVIDIA Visual Profiler, and NVIDIA Nsight Compute is a plus • Familiar with LLM inference optimization, experience in developing popular inference framework such as TensorRT-LLM, SGLang, …
工作职责
• Lead the software development in C/C++, Python, and in GPU languages such as CUDA, ROCm, or Triton• Analyze metrics and identify opportunities based on offline and online testing, develop and deliver robust and scalable solutions.• Work with cutting-edge hardware stacks and a fast-moving software stack to deliver best-of-class inference and optimal cost.• Engage with key partners to understand and implement inference and training optimization for state-of-the-art LLMs and other models.
• Own the design and implementation of core infrastructure components, ensuring scalability, security, and high performance across systems. • Lead architectural decisions, set engineering standards, and drive long-term technical strategy for complex solutions. • Produce and review high-quality, secure, and maintainable code while championing best practices and modern patterns, including AI-driven approaches. • Mentor engineers across teams, fostering growth in coding, design, testing, and operational excellence. • Define and enforce robust test strategies, integrate automation, and ensure security and reliability in all deliverables. • Oversee telemetry, incident response, and operational readiness to improve system stability and supportability. • Collaborate with stakeholders to validate requirements, incorporate feedback, and uphold compliance, privacy, and accessibility standards.
- Keep up to date with and utilize the latest developments in LLM system optimization.- Take the lead in designing innovative system optimization solutions for internal LLM workloads.- Optimize LLM inference workloads through innovative kernel, algorithm, scheduling, and parallelization technologies.- Continuously develop and maintain internal LLM inference infrastructure.- Discover new LLM system optimization needs and innovations.
• Develop secure, performant, and maintainable code across products; mentor teams on best practices and validation strategies while ensuring adherence to design specifications. • Own complex product architecture decisions, evaluate design options, and ensure scalability, resiliency, compliance, and cost-effectiveness. • Define and implement robust test strategies, including security testing; improve test coverage, reliability, and automation leveraging AI tools. • Build and release thoughtful features for an engaging homepage experience—primarily frontend (full stack preferred)—across multiple products (e.g., Edge browser, Windows Widgets) and endpoints (desktop, mobile). • Promote secure design principles, scalability, and performance; stay current with emerging technologies and collaborate closely with developers and PMs across countries for agile development. • Integrate telemetry for monitoring system health, lead incident response and postmortems, and proactively improve live-site operations and troubleshooting processes. • Partner with stakeholders to validate customer needs, incorporate feedback loops, and ensure solutions meet security, privacy, and usability expectations.
• Partners with appropriate stakeholders to determine user requirements for a set of scenarios. • Leads identification of dependencies and the development of design documents for a product, application, service, or platform. • Leads by example and mentors others to produce extensible and maintainable code used across products. • Leverages subject-matter expertise of cross-product features with appropriate stakeholders (e.g., project managers) to drive multiple group's project plans, release plans, and work items. • Holds accountability as a Designated Responsible Individual (DRI), mentoring engineers across products/solutions, working on-call to monitor system/product/service for degradation, downtime, or interruptions. • Proactively seeks new knowledge and adapts to new trends, technical solutions, and patterns that will improve the availability, reliability, efficiency, observability, and performance of products while also driving consistency in monitoring and operations at scale and shares knowledge with other engineers.