Senior Software Engineer

Microsoft Silicon, Cloud Hardware, and Infrastructure Engineering (SCHIE) is the team behind Microsoft’s expanding Cloud Infrastructure and responsible for powering Microsoft’s “Intelligent Cloud” mission. SCHIE delivers the core infrastructure and foundational technologies for Microsoft's over 200 online businesses including Bing, MSN, Office 365, Xbox Live, Teams, OneDrive, and the Microsoft Azure platform globally with our server and data center infrastructure, security and compliance, operations, globalization, and manageability solutions. Our focus is on smart growth, high efficiency, and delivering a trusted experience to customers and partners worldwide and we are looking for passionate engineers to help achieve that mission.

Microsoft's Hardware Systems organization is developing AI-native silicon and hyperscale systems designed to power the next generation of frontier AI models. The MAIA platform combines custom silicon, high-performance networking, advanced compiler technologies, and large-scale system infrastructure to enable industry-leading AI training and inference.

The Platform Systems Engineering (PSE) team is seeking a AI Accelerator Tools Development Engineer to lead the development of next-generation stress, validation, and performance tooling for MAIA AI accelerator platforms.

In this role, you will build software frameworks, stress workloads, and validation tools that exercise every layer of the AI stack, from hardware execution engines and memory subsystems to compiler-generated kernels, distributed communication fabrics, and large-scale AI workloads. Your work will play a critical role in platform bring-up, qualification, performance characterization, reliability validation, and fleet readiness for both current and future generations of MAIA systems.

You will work closely with silicon architects, compiler teams, runtime developers, performance engineers, validation teams, and AI framework developers to translate platform requirements into scalable tooling and workload solutions.

#azure #MAIA #AI/ML

AI Workload & Stress Tool Development

• Design and develop scalable stress, performance, and validation frameworks for MAIA AI accelerator platforms.
• Build workload generation infrastructure capable of exercising compute, memory, interconnect, networking, storage, and system-level resources.
• Develop reusable stress tools using PyTorch, Triton, Python, C++, and custom MAIA SDKs.
• Create synthetic and production-inspired workloads that model training and inference behaviors observed in large-scale AI deployments.
• Build automated infrastructure for workload deployment, orchestration, telemetry collection, and result analysis.

Hardware-Aware Workload Optimization

• Develop and optimize kernels targeting custom AI accelerators.
• Create GEMM, attention, collective communication, and memory intensive stress workloads.
• Analyze execution behavior across the hardware-software stack and identify bottlenecks impacting utilization and performance.
• Collaborate with compiler and runtime teams to improve workload efficiency and hardware utilization.

Compiler & SDK Integration

• Develop tooling that integrates with MAIA compiler pipelines, SDKs, runtime environments, and performance analysis tools.
• Understand and debug compiler output, generated kernels, scheduling decisions, and execution behavior.
• Build automation around model compilation, kernel validation, regression testing, and workload portability.
• Partner with compiler teams to validate new compiler features and workload optimization strategies.

Platform Validation & Reliability

• Design workload suites for platform bring-up, qualification, and reliability testing.
• Build comprehensive regression infrastructure supporting silicon, firmware, system software, and platform releases.
• Develop automated validation tools capable of identifying correctness, performance, thermal, power, and stability issues.
• Enable platform readiness through scalable validation methodologies and continuous regression testing.

Performance Engineering

• Characterize system performance across compute, networking, memory, and storage subsystems.
• Develop benchmarking methodologies and performance dashboards.
• Adapt and optimize industry-standard workloads including:
• HPL/HPC benchmarks
• LLM training workloads
• Transformer-based inference workloads
• Collective communication benchmarks
• AI framework benchmark suites
• Drive root-cause analysis and optimization initiatives across the stack.

Developer Productivity & Automation

• Improve developer productivity through automation, CI/CD integration, diagnostics, and debugging infrastructure.
• Build reusable tooling for workload generation, failure triage, telemetry analysis, and reporting.
• Develop dashboards and automated workflows for large-scale validation environments.
• Partner with engineering teams to convert recurring validation challenges into durable tooling solutions.

Software Engineering

• BS/MS in Computer Science, Computer Engineering, Electrical Engineering, or equivalent experience.
• 8+ years of software development experience.
• Strong programming skills in:
• Python
• C/C++
• Experience designing production-quality software systems and frameworks.

AI & Accelerator Experience

• Experience developing workloads for GPUs, AI accelerators, or HPC systems.
• Deep understanding of AI training and inference workloads.
• Hands-on experience with:
• PyTorch
• Triton
• Distributed AI workloads
• Experience developing or optimizing compute-intensive kernels.

Hardware Systems Knowledge

• Understanding of modern accelerator architectures, including:
• Compute engines
• Memory hierarchy
• Interconnects
• Runtime systems
• Experience with performance profiling, bottleneck analysis, and workload optimization.
• Familiarity with distributed systems and large-scale AI infrastructure.

Validation & Performance

• Experience developing stress, benchmark, validation, or reliability workloads.
• Experience building automated test frameworks and regression infrastructure.
• Strong debugging and root-cause analysis skills across hardware and software boundaries.

Preferred Qualifications:

• Experience working with custom AI accelerator SDKs and compiler ecosystems.

• Familiarity with compiler technologies such as:

• LLVM, MLIR & Triton Compiler

• Experience with kernel generation frameworks and code generation pipelines.
• Experience with LLM training, inference, and AI model optimization.
• Understanding of collective communication libraries and networking technologies.
• Experience with performance characterization of large-scale AI clusters.
• Familiarity with CI/CD systems, containerized environments, and cloud-scale validation infrastructure.
• Experience working with silicon bring-up or post-silicon validation teams.

Ability to meet Microsoft, customer and/or government security screening requirements are required for this role. These requirements include but are not limited to the following specialized security screenings: Microsoft Cloud Background Check: This position will be required to pass the Microsoft Cloud Background Check upon hire/transfer and every two years thereafter.

This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled.

Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, citizenship, color, family or medical care leave, gender identity or expression, genetic information, immigration status, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran or military status, race, ethnicity, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable local laws, regulations and ordinances. If you need assistance with religious accommodations and/or a reasonable accommodation due to a disability during the application process, read more about requesting accommodations.