Software Engineer II - AI/ML Infrastructure CoreAI

If you’re excited by the challenge of shaping the future of AI, the AI Platform organization at Microsoft CoreAI offers you the chance to make a real impact. As part of the Azure AI Foundry Customization team, you’ll help build the cutting-edge platform that powers Azure’s innovation—and all of Microsoft’s flagship products, from GitHub and Microsoft 365 to Xbox. Our mission is simple but ambitious: to empower every AI application on the planet.

Here, you’ll work at the heart of a deeply technical environment, designing systems that enable AI to learn, improve, and scale in production. You’ll tackle some of the toughest industry challenges — making frontier models not just powerful, but also accessible, efficient, and cost-effective for enterprises everywhere. By developing pre-training, mid-training, and post-training solutions, you’ll help organizations build AI systems that continuously learn and adapt, using the models most relevant to their workloads.

This role gives you end-to-end ownership and exposure, from high-level APIs to granular reinforcement learning controls, and from Azure ML flexibility to deploying LoRA models and inferencing at scale. You’ll deliver services, pipelines, ML recipes, evaluations, and integrations—supported by Azure’s world-class data centers and GPU infrastructure. With innovations like single-click training data generation, dynamic model routing, and a global partner ecosystem, you’ll help close the loop between deployment and learning, unlocking massive improvements in cost, speed, and reliability.

Join us and you’ll have the opportunity to showcase your work on platforms like Microsoft Build and Microsoft Ignite, where we highlight breakthroughs in industry-leading innovation, like these (from June 2026):

• Deploy. Observe. Learn. Reinforcement learning for production agents

• Turn foundation models into production AI on Microsoft Foundry

• Post-Training and Deploying Open Source Reasoning Models in Foundry

• Smaller, faster, smarter: Distilling models with fine‑tuning

• Microsoft Build opening keynote for all the exciting innovation @ Microsoft

If you’re ready to shape the future and make agentic AI economically viable for every enterprise, this is where your journey begins.

As a contributor in a customer-obsessed team, you will collaborate with data scientists and other developers on the team to create the necessary abstractions, infrastructure, interfaces and implement features to train, test, validate, scale and optimize for machine learning algorithms including but not limited to data cleaning, model training, hyper parameter tuning at massive, distributed scale. You will drive the incorporation of these customer-inspired innovations into our software. You will take holistic ownership of your features.

Your responsibilities will include:

• Independently use appropriate AI agents, skills, and practices across the software development lifecycle.

• Work with appropriate internal stakeholders to understand customer/user requirements for a set of features

• Create a clear test strategy that ensures solution quality, executes test plans, and builds testable code. 

• Create extensible and maintainable product feature code for minimal defects. 

• Review product feature code to ensure it contains the correct test coverage and follows team standards. Implements debugging tools, tests, logs, and telemetry to verify assumptions.

• Strive for automation within production deployment aiming for zero-touch when possible. 

• Apply security best practices to achieve security hardening across product features.

Qualifications 

(Required and Preferred)