Tag
The TiDB team deliberately avoids using Multi-Agent systems internally, based on lessons learned from distributed systems and referencing Anthropic's related research, emphasizing that optimizing a single Agent might be more effective.
This tweet discusses the challenges of building LLM Agent memory on top of a filesystem, cites a related paper, and introduces the design philosophy of TiDB Filesystem.
TiDB in the AI era is gradually evolving from a database into an agent-oriented unified storage layer, a natural evolution driven by customer demand.