Date: June 25, 2026
Time: 10:00am – 11:00am PT
For two decades, the largest SaaS workloads have pushed databases past the limits of single-system architectures. Companies including Atlassian built their growth on architectures designed to absorb millions of users, with millions of database/tables, and unpredictable workload spikes. Now agentic AI is testing those same limits, at higher frequency and with new workload shapes: Long-running agent sessions, persistent memory across millions of concurrent agents, and unified vector and relational queries that traditional database stacks were never designed to serve.
An analysis of more than 60 production TiDB deployments points to a single recurring theme: Customers had scaled into complexity. Cluster sprawl, stitched-together database products, and operational overhead that compounded with every new workload. The move that worked for them, consolidating complexity into a unified data layer, is the same move teams building for production AI are being forced to make.
In this webinar, TiDB’s Field CTO and Principal Architect, Terry Purcell, takes technical leaders through the architectural evolution that has carried massive-scale SaaS customer use cases into the agentic AI era now being defined by companies like Manus. Drawing on decades of enterprise database experience, Terry walks through the consolidation patterns that have separated systems that scale through each era from those that have to be replaced.
What you’ll learn:
- The architectural arc from massive-scale SaaS workloads to production agentic AI, and what’s actually different this time.
- How TiDB customers consolidated complexity at their scale inflection points.
- How incremental automation, including runbooks, SDKs, and async job runners, compounds into a decisive speed advantage.
- Where agentic workloads push database systems beyond previous-era limits, including object scale (millions of tables, thousands of columns) and elastic capacity.
- Why the consolidation pattern that worked for SaaS scale is the same one production AI now needs.
- Real-world lessons from production AI deployments, including companies like Manus.
Speakers:
Terry Purcell is Field CTO and Principal Architect at TiDB.
He brings decades of enterprise database experience to the role, spanning query optimization, distributed systems performance, and large-scale enterprise database architecture. At TiDB, he advises organizations on the data infrastructure decisions behind production AI workloads, helping technical leaders evaluate where traditional systems break down under agentic AI traffic and what architectural patterns hold up at scale.