3 days ago
Remote, Japan or Tokyo, JapanStaff+
Responsibilities
- Drive improvements to end-to-end inference performance for OpenAI and other state-of-the-art LLMs.
- Benchmark and measure model performance on NVIDIA and AMD GPUs and Microsoft silicon.
- Optimize and monitor LLM performance and build software tooling for model- and system-level performance insights.
- Build tools that accelerate porting and deploying models on new NVIDIA and AMD GPUs.
- Design, implement, and test components for AI, DNN, and LLM frameworks and tools.
- Reduce the complexity and improve the performance and efficiency of key system components and pipelines.
- Collaborate with internal and external partners, researchers, and developers.
Requirements
- Bachelor's degree in computer science or a related technical field and 4+ years of technical engineering experience with coding, or equivalent experience.
- Experience coding in languages including C, C++, C#, Java, JavaScript, or Python.
- Ability to meet Microsoft, customer, and government security screening requirements, including the Microsoft Cloud Background Check.
- Preferred: master's degree with 8+ years of technical engineering experience, or bachelor's degree with 12+ years of experience, or equivalent experience.
- Technical foundation in software engineering, computer architecture, GPU architecture, and hardware neural-network acceleration.
- Experience analyzing and optimizing end-to-end performance of state-of-the-art LLMs and using GPU profiling tools.
- Experience with DNN/LLM inference and one or more deep-learning frameworks such as PyTorch, TensorFlow, or ONNX Runtime, plus familiarity with CUDA, ROCm, or Triton.
- Ability and desire to collaborate across teams of researchers and developers.
Tech Stack
Categories
About Microsoft
Microsoft develops operating systems, productivity software, cloud services, developer tools, and consumer devices for individuals, enterprises, and governments. Its main products include Windows, Microsoft 365, Azure, Visual Studio/GitHub, Xbox, and LinkedIn; revenue comes from software subscriptions and licenses, cloud consumption, hardware sales, and advertising. Founded in 1975 and headquartered in Redmond, Washington, Microsoft is a public company traded on Nasdaq.