Develop faster on Windows with AI playbooks and local agents | DEMSP386
Adrian Macias presents a practical “setup-to-ship” workflow for developing on Windows using AI playbooks and local agents on AMD Ryzen AI PCs, focusing on reducing friction from idea to first commit and showing how local execution can scale across Windows systems optimized for AMD hardware.
Overview
The session walks through a streamlined developer workflow that emphasizes local coding and agent-driven assistance on Windows systems optimized for AMD hardware.
Key topics covered
Emerging AI use cases
- AI-driven automation scenarios
- Gaming-related AI use cases
- Coding agents and agentic workflows
Locality and workload distribution
- How workloads can be split between client and cloud
- Considerations for when to run locally versus in the cloud
Offline-capable development workflow
- A demonstration of mobile development and debugging without cloud connectivity
Model routing across hardware accelerators
- Routing AI workloads across CPU, GPU, and NPU
- Choosing execution targets based on system context
Frameworks and model orchestration concepts
- An introduction to the Lemonade framework and its community-driven development approach
- An “omni modal” capability that combines multiple AI models
- A vLLM router concept and an agentic semantic filtering use case
Local vs. cloud application use cases
- Discussion of scenarios where local execution is preferred
- Discussion of scenarios where cloud execution is preferred
Session metadata
- Event: Microsoft Build 2026
- Session code: DEMSP386
- Level: Intermediate
- Track: Developer tools & frameworks