Microsoft Build 2026: Lightning Talk + Product Booth

Location: San Francisco, CA (Microsoft Build)
Role: Lightning talk speaker; Redis Expert; Azure Redis booth duty
Audience Size: 15-minute lightning talk (30 attendees) + 1:1 booth conversations

I spoke at Microsoft Build 2026 in a 15-minute lightning talk on building faster, smarter agents with Azure Managed Redis and Foundry. The session focused on a simple idea: Foundry orchestrates the agent, Redis gives it memory.

The talk covered three patterns: semantic caching, vector search for RAG, and persistent agent memory. I showed how Azure Managed Redis can sit in the hot path of an agent to reduce repeated model calls, retrieve grounded context, and preserve memory across sessions.

I also supported the Azure Redis booth as a Redis Expert, speaking with customers, partners, Microsoft field teams, and product stakeholders about AMR adoption, agent architectures, semantic caching, vector search, and migration from Azure Cache for Redis.

(Topics covered in the talk):

• Stateless agents: how to avoid agents repeatedly re-fetching context and re-calling the model.

• Semantic caching: using Redis to cache responses by meaning, not just exact prompt text.

• Vector search for RAG: retrieving ranked, source-backed context before the model generates an answer.

• Persistent memory: separating short-term conversation state from long-term user preferences and learned facts.

• Foundry integration: using Azure Managed Redis as the memory layer alongside Microsoft Agent Framework and Foundry Models.

• Production readiness: lowering latency and token spend while making agents more useful across sessions.

• Delivered a Microsoft Build lightning talk on Azure Managed Redis for Foundry agent architectures.

• Showed how Redis supports semantic caching, vector RAG, and persistent memory in one agent memory layer.

• Supported Azure Redis booth duty as a Redis Expert, helping attendees think through real deployment questions.

• Engaged customers, partners, Microsoft field teams, and product stakeholders on AMR use cases, migration paths, and AI agent patterns.