Build AI across client and cloud with AMD ROCm and Microsoft | BRKSP93

Author: Microsoft Developer - Bewertung: 4x - Views:48

As AI moves from experimentation to production, developers need practical ways to build, test, and optimize workloads across client, cloud, and on-prem environments. Learn how AMD ROCm™ provides a common AI software foundation across Radeon™, Ryzen™ AI, and AMD Instinct™ platforms. See how integrations with PyTorch and ONNX/MIGraphX enable portability with fewer code changes while optimizing performance for each target.



Seating for this session is first-come, first-served. Add it to your schedule to plan your day and arrive early to secure a spot.



𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀:

* Anush Elangovan





𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻:

This is one of many sessions from the Microsoft Build 2026 event. View even more sessions on-demand and learn about Microsoft Build at https://build.microsoft.com



BRKSP93 | English (US) | Developer tools & frameworks





Breakout | (300) Advanced





#MSBuild



Chapters:

0:00 - Focus on execution velocity as the key advantage in modern technology

00:02:30 - Shifting value from syntax implementation to intent-driven execution

00:03:21 - Parallel Development and Accelerated Feedback Loop

00:04:09 - Building a Unified and Pervasive Software Layer

00:10:07 - Enhanced Model Support and Memory Capabilities on AMD Platforms

00:12:04 - Explanation of client application architecture and proxy workflow

00:13:49 - Description of loaded models for domain classification, jailbreak, and PII detection

00:15:00 - Handling private data routed to local model for security

00:15:46 - Configuration flexibility for routing queries between models