Build and deploy local AI at scale across enterprise PCs w/ Windows ML | BRK388
Anastasiya Tarnouskaya, Pawan Tahiliani, and Daniel Freedman discuss Windows ML’s role in running AI locally across a diverse enterprise PC fleet, with a focus on deployment at scale.
Local inference across enterprise hardware
Moving custom and open-source AI models from a pilot into production means supporting PCs with different hardware capabilities. This session examines how Windows ML streamlines deployment and provides a single framework for high-performance local inference across CPU, GPU, and NPU hardware.
The focus is on two connected challenges:
- Scaling model deployment beyond a pilot to a broader enterprise PC fleet.
- Using one framework to support local inference across different types of compute hardware.
Session details and resources
- Event: Microsoft Ignite 2026
- Session: BRK388 | Windows
- Format: Breakout
- Level: (400) Expert
Explore the Windows next-step resources provided for the session.