Principal Software Engineer - Compute Graph Compiler
About The Position
The Compute Graph Compiler (CGC) team within DirectX builds compiler and execution technologies for the Windows Machine Learning Platform. Its mission is to enable efficient machine-learning workload execution across CPUs, GPUs, and NPUs from a broad ecosystem of silicon partners. CGC technologies support computer vision, audio, video, gaming, and generative AI workloads, including emerging AI-first experiences and Copilot+ PC capabilities. The team develops compiler infrastructure, optimization frameworks, execution engines, and hardware abstractions, working with Microsoft AI platform teams, hardware partners, and operating system components to advance AI execution on Windows. The Principal Software Engineer will define and drive the next generation of AI compute infrastructure for Windows, leading architecture and technical strategy across compiler technologies, execution runtimes, and hardware acceleration platforms. The role involves influencing Microsoft's AI platform direction, collaborating with silicon partners, and developing technologies for efficient execution of increasingly complex AI workloads.
Requirements
- Bachelor's degree in Computer Science or a related technical field and 6+ years of technical engineering experience with production-quality systems software, AI/ML platforms, or related technologies, or equivalent experience.
Nice To Haves
- Demonstrated experience designing and delivering production-quality systems software, AI/ML platforms, or related technologies.
- Hands-on experience with ML compilers, runtimes, frameworks, tools, and platform technologies.
- In-depth operating system development experience, including user-mode and/or kernel-mode driver development.
- Experience designing ML accelerators or hardware microarchitecture for real-world products and connecting that hardware with compilers, runtimes, drivers, and frameworks.
- Knowledge of computer architecture, heterogeneous computing, memory systems, parallel processing, and hardware/software co-design.
- Proven experience optimizing ML workloads across the full software and hardware stack, including GPU optimization.
- A track record of technical leadership defining architecture and driving complex AI/ML research and development across hardware and software teams, including taking AI platforms or accelerators from early design and prototypes through production.
Responsibilities
- Define and drive the technical vision, architecture, and roadmap for key components of the Windows ML compiler stack.
- Lead cross-team initiatives to improve AI workload performance, scalability, and efficiency across CPU, GPU, and NPU hardware.
- Drive NPU enablement strategy and advanced optimization techniques for next-generation AI processors.
- Partner with hardware vendors and internal platform teams to influence future AI hardware and software capabilities.
- Design and develop high-performance compiler infrastructure, execution engines, and developer tooling.
- Identify and solve complex systems challenges involving scheduling, memory management, code generation, graph optimization, and hardware abstraction.
- Mentor engineers and provide technical leadership across multiple projects and organizations.
- Help shape Microsoft's long-term strategy for on-device AI and generative AI execution on Windows.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Principal
