Principal Software Engineer - AI Infrastructure

Help build the infrastructure that trains the next generation of agentic Artificial Intelligence (AI) models at Microsoft. The Cloud + AI (C+AI) organization builds the end-to-end AI stack that powers innovation across Azure and Microsoft. The Foundry, Inference and Training (FIT) Training Team is where frontier-scale model training meets production-grade engineering. You will work alongside engineers and researchers to advance the scientific methods and scalable infrastructure used to train Large Language Models (LLMs), Small Language Models (SLMs), and agentic models, spanning both proprietary and open-weight models and building on open-source training frameworks. Your work will directly shape how reliable, enterprise-grade agentic workflows are built across the company and experienced by the customers who depend on them.

As a Principal Software Engineer on this team, you will own the architecture and technical direction of major components of the training platform and drive them from ambiguous problem statements through production operation. You will keep the customer at the center of every decision, partnering directly with the researchers and product teams who rely on the platform to understand their needs, remove friction, and measure success by their outcomes. You will set the technical bar for the team through design leadership, code and test-strategy reviews, and mentorship, and you will partner across C+AI and with partner organizations to align infrastructure roadmaps with model training and evaluation needs. This opportunity will allow you to deepen your expertise in large-scale distributed systems for AI training, develop firsthand understanding of how frontier proprietary and open-weight models are trained and evaluated, and grow your influence as a technical leader who shapes architecture, engineering practices, and the next generation of engineers on a fast-moving, startup-style team.

We are looking for a curious, independent, adaptable, and customer-obsessed problem-solver who thrives on continuous learning, is energized by evolving priorities, and is motivated by the impact your work has on the people who use it. You write efficient, well-tested code, enjoy debugging hard problems in complex distributed systems, and share what you learn. You iterate quickly, adjust course when the evidence calls for it, and collaborate effectively in a dynamic environment.

Microsoft’s mission is to empower every person and every organization on the planet to achieve more. As employees we come together with a growth mindset, innovate to empower others and collaborate to realize our shared goals. Each day we build on our values of respect, integrity, and accountability to create a culture of inclusion where everyone can thrive at work and beyond

Responsibilities

• Own the architecture and technical direction of major components of the training platform for Large Language Models (LLMs), Small Language Models (SLMs), multimodal, code-specific, and open-weight models, ensuring designs meet scalability, reliability, security, and cost-efficiency requirements.
• Partner directly with customers, including researchers and product teams across Microsoft, to understand their training and evaluation needs, translate them into platform requirements, and measure success through customer experience and outcomes.
• Drive ambiguous, multi-quarter initiatives from proposal to production, breaking work into milestones, identifying dependencies, and negotiating ownership and delivery commitments with partner teams across Cloud + AI (C+AI).
• Build on and extend open-source training frameworks (for example, PyTorch, DeepSpeed, Megatron-LM, and Hugging Face libraries), adapting them for proprietary and open-weight models and contributing improvements back to the community where appropriate.
• Design and build services that serve production traffic while meeting Microsoft security, privacy, Responsible AI, and compliance requirements, including auditable evidence of compliance.
• Lead debugging and resolution of complex issues in distributed training jobs and production services; participate in the on-call rotation as a Designated Responsible Individual (DRI), lead incident retrospectives, and improve troubleshooting guides, testing, safe deployment, and monitoring.
• Mentor engineers and raise the technical bar through design reviews, code reviews, and ownership of the team's test strategy, while championing AI-assisted engineering tools and practices across the software development lifecycle.

Qualifications

Required Qualifications:

• Bachelor's Degree in Computer Science or related technical field AND 6+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python OR equivalent experience.

Preferred Qualifications:

• Master's Degree in Computer Science or related technical field AND 8+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python OR Bachelor's Degree in Computer Science or related technical field AND 12+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python OR equivalent experience.

• 5+ years designing, building, and operating large-scale distributed systems or cloud platforms with end-to-end ownership of production services, including Kubernetes-based containerized platforms on Microsoft Azure, Amazon Web Services (AWS), or Google Cloud, control-plane or backend services that manage compute, storage, and networking, and secure multi-tenant designs covering authentication, authorization, workload identity, quota management, and isolation.
• 3+ years building machine learning training infrastructure, such as Graphics Processing Unit (GPU) cluster scheduling, data pipelines, and checkpointing and storage systems, using open-source training frameworks such as PyTorch Fully Sharded Data Parallel (FSDP), DeepSpeed, Megatron-LM, Hugging Face Transformers and Accelerate, or Ray Train; contributions to open-source projects are a strong plus.
• 2+ years training, fine-tuning, or evaluating open-weight models (for example, Llama, Mistral, Phi, Qwen, or DeepSeek model families) at multi-node scale.
• 2+ years building or operating Reinforcement Learning (RL) post-training pipelines for Large Language Models (LLMs) and agentic models using verl, slime, MILES, or comparable frameworks, with a demonstrated understanding of the end-to-end training stack, including pretraining and post-training frameworks, Reinforcement Learning Environments (RLEs), rollout inference engines (for example, vLLM or SGLang), evaluation, and checkpointing, and how design choices at each layer affect throughput, stability, and cost.
• 3+ years serving as an on-call or Designated Responsible Individual (DRI) engineer for production services operating against defined availability or latency targets, including incident response, postmortems, structured logging, distributed tracing, dashboards, alerting, automated validation, and incident playbooks.
• 2+ years partnering directly with internal or external customers to define requirements, instrument customer experience (for example, adoption, satisfaction, or time-to-first-result metrics), and prioritize engineering work by customer impact.
• Experience leading technical design across 2+ teams, driving ambiguous multi-quarter initiatives to completion, and mentoring senior engineers.

#AIPLATFORM #AIINFRA

Software Engineering IC5 - The typical base pay range for this role across the U.S. is USD $142,800 - $274,800 per year. There is a different range applicable to specific work locations, within the San Francisco Bay area and New York City metropolitan area, and the base pay range for this role in those locations is USD $188,000 - $304,200 per year.

Certain roles may be eligible for benefits and other compensation. Find additional benefits and pay information here:

https://careers.microsoft.com/us/en/us-corporate-pay

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.