Why Your Enterprise Tech Stack Isn’t Ready for AI Agents — Christopher Lovejoy & Saul Howard
Source
youtube.com
Author
AI Engineer
Date
Context
An AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition →'s action history can be recorded separately from the sensitive records it used. In Christopher Lovejoy and Saul Howard's proposed architecture, an append-only event log records actions, data access and authorization, while patient information lives in separate object storage. Events contain references to those records rather than copies of their contents.
The intended benefit is that engineers can investigate an agent's sequence of actions without routinely viewing patient information. Recording events as they happen also avoids reconstructing the history afterward. The speakers acknowledge a storage tradeoff: preserving an immutable history makes reading it more complicated.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.