chDB as the Agent's Local Data Engine

Summary

Agents often make 5 to 20 tool calls per turn, and many of those calls retrieve data such as prior memory, recent events, or catalog records. When data is remote, round-trip latency compounds across sequential calls, while packet loss can trigger retries that increase both task time and token usage. The article presents chDB, which embeds a full ClickHouse query engine in the agent process, as a way to make data access local and CPU-bound rather than network-bound. For local memory, it combines structured, time-series, vector, and text-index capabilities with local storage and an embedding model, while append-only MergeTree rows preserve belief revisions, deletion markers, and point-in-time history. It also describes syncing trickle-written local memory to a warehouse on a slower cadence and using a materialized-view mirror for frequently accessed current state.

Takeaways

1. 1

   Agent memory is modeled as queryable data rather than simply a larger prompt; chDB can combine structured, time-series, vector, and inverted text-index capabilities in one local engine.

2. 2

   The append-only MergeTree design preserves belief evolution: each revision is an immutable row, while is_deleted marks removals without erasing history. Queries can return current state, full history, or point-in-time beliefs.

3. 3

   Local memory can absorb small, continuous writes without pressuring an online warehouse, then synchronize those writes on an hourly or daily archival cadence.