Senior Software Engineer - Maia Model Enablement

Microsoft • ,

7h • $119,800 - $261,000

About The Position

The Maia Model Enablement team owns the end-to-end technical work required to bring frontier AI models onto Microsoft's AI accelerators and make them successful in production. This work spans model enablement, performance optimization, serving capabilities, validation infrastructure, hardware bring-up, and cross-stack debugging.

The Senior Software Engineer solves end-to-end systems challenges at the intersection of AI models, software, and hardware. The role works across the software stack, from model architectures and serving systems to infrastructure and hardware interactions, and adapts to changing problems throughout the lifecycle of models and hardware platforms. The team relies on measurement, experimentation, automation, and AI-assisted engineering workflows, and partners closely with model, software, and hardware teams to deliver production-ready solutions.

Success includes enabling new model architectures to run correctly on Maia; meeting production performance, scalability, and reliability targets; resolving bottlenecks throughout the stack; making new hardware generations usable by model teams; and making model onboarding faster, more repeatable, and less operationally intensive. The position requires a bachelor's degree in Computer Science or a related technical field and 4+ years of technical engineering experience coding in C++, Python, or equivalent experience. The role also requires passing Microsoft's Cloud Background Check upon hire or transfer and every two years thereafter.

Requirements

• Bachelor’s Degree in Computer Science or a related technical field and 4+ years of technical engineering experience coding in languages such as C++, or Python, or equivalent experience.
• Ability to meet Microsoft, customer and/or government security screening requirements.
• Must pass the Microsoft Cloud Background Check upon hire or transfer and every two years thereafter.
• Proficiency in software development with languages such as Python, C++, or equivalent technologies.

Nice To Haves

• Understanding of modern AI model architectures, inference systems, and performance characteristics.
• Experience working with AI/ML frameworks, runtimes, or serving stacks such as PyTorch, ONNX Runtime, vLLM, TensorRT-LLM, or similar technologies.
• Experience diagnosing and resolving complex issues across software, systems, and hardware layers.
• Familiarity with performance analysis, distributed systems, accelerator architectures, or large-scale production services.
• Experience enabling software on new hardware platforms, accelerators, or large-scale systems through activities such as bring-up, performance characterization, hardware-software debugging, or platform enablement.
• Experience using AI-assisted engineering tools, agents, skills, and workflow automation to accelerate software development, debugging, code maintenance, investigation, and engineering productivity.

Responsibilities

• Bring new AI models and model architectures to Maia.
• Diagnose and resolve issues across software, systems, and hardware layers.
• Improve performance, scalability, and serving capabilities for production AI workloads.
• Build automation, validation infrastructure, and engineering systems that accelerate model onboarding.
• Partner across model, runtime, infrastructure, and hardware teams to remove obstacles and deliver production impact.
• Work across the software stack, from model architectures and serving systems to infrastructure and hardware interactions.
• Use measurement, experimentation, automation, and AI-assisted engineering workflows to move quickly and scale impact.
• Enable new model architectures to run correctly on Maia and help production models achieve performance, scalability, and reliability targets.
• Identify and resolve bottlenecks across the stack, help make new hardware generations usable by model teams, and improve model onboarding.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior