Building the Data Layer for AI Agents

AI has crossed from demos into production to agents that remember, reason, and act over time. That shift moves the hard problem off the model and onto the data layer. Durable memory, real-time context, retrieval, and per-tenant isolation now have to work at scale, in one system. Spend an evening with AWS and TiDB working through that shift together. AWS brings the platform and the market view. TiDB brings the data layer built for this problem: distributed SQL that scales without sharding, HTAP on live data, and native vector search with durable state for agent memory, running on AWS alongside Bedrock and AgentCore. One foundation carries the transactional scale you need today and the agentic scale you’ll need tomorrow, so you don’t re-platform to get from one to the other. Moonshot AI’s Kimi proves it in production, running tens of millions of isolated agent tenants on TiDB with each one provisioned in under a second. What the frontier does first, the market does next. You’ll leave with a market map of where AI value is real, an architecture you can apply to your own workload, and a read on the demand curve heading your way. Programme 17:00 — Arrival, drinks, and networking17:30 — Welcome and framing17:35 — Topic and Speaker to be announced, AWS17:55 — The Agentic Data Layer on AWS: TiDB and AWS, Architecturally. Bernard Kavanagh, TiDB18:15 — Lessons from the Frontier: Inside the Kimi Build, and What APAC Signals Next. Siddon Tang, TiDB18:35 — Fireside Q&A with all three speakers18:50 — Beer and finger food Speakers: Speaker to be announced AWS Bernard Kavannagh, Principal Solutions Architectm Applied AI, TiDB Bernard is a Principal Solutions Architect focused on Applied AI, designing production architectures for agentic and AI-native workloads. He turns the TiDB + AWS joint value into reference patterns engineering teams can actually deploy — collapsing the transactional database, vector store, cache and analytics stack into one engine on AWS, and showing where it fits for payments, multi-tenant SaaS, real-time risk and agentic features. Siddon Tang, Chief AI Officer & GM, APAC, TiDB Siddon is one of the foundational engineers behind TiDB and the creator and long-time lead of TiKV, its distributed storage engine and a CNCF-graduated project. Across a decade building TiDB he has become one of the most recognized voices in distributed databases and the Rust systems community. As Chief AI Officer and GM of APAC he now leads TiDB’s AI direction and its work with frontier AI companies — including Moonshot AI’s Kimi agent-hosting platform, one of the largest agentic workloads in the world