Building a Knowledge Base That LLMs Can Actually Search

Documentation written by the team that builds the system tends to be more useful than documentation written by anyone else. The trade-off is consistency, which we address with a shared style guide and a lightweight review process.

What we changed

Tool definitions should read like API documentation written for a careful junior engineer. The model behaves better when each parameter has a concrete example, a unit, and an explicit statement of what happens when the value is omitted.

Hardware is a moving target. The Jetson Orin we benchmarked in January was outperformed by an off-the-shelf mini-PC by August. We re-run the benchmark matrix every quarter and have stopped making long-term hardware commitments.

Next steps

Observability for agent runs is qualitatively different from traditional APM. A single user request can spawn dozens of tool calls, each with its own latency, cost, and failure mode. Flat traces become unreadable; we render them as collapsible trees.

When the system is wrong, the user should be able to understand why in under thirty seconds. Citation links, confidence scores, and the exact retrieved passages are surfaced in the UI for every generated answer.

In production, latency distributions matter far more than averages. A pipeline whose mean response time looks acceptable can still feel sluggish if the 95th percentile drifts upward during peak hours. We instrument every stage with histograms so regressions surface immediately.