Stefania Druga treats agent memoryAI agent memory is stored information that an agent can retrieve and use across steps, sessions, or changing contexts. as a write, manage and read control loop rather than a database alone. Her harnessAn AI agent harness is the software framework that packages a model with tools, instructions, context management, execution controls, and user interaction. keeps a core trace, tests several recall policies and uses an oracle that supplies ground-truth memory without forcing the model to use it.
On a literature-review task that fit within the model context, memory did not improve performance and only added cost. On XBench questions whose answers fell hundreds of steps outside the active context window, ranked recallRetrieval is the process of selecting relevant stored information and returning it to an AI system for the current task. improved answer accuracy over no memory and simple gating.
The best-performing setup ranked a decision ledger, and the result held across both tested local models and an additional benchmarkA benchmark is a defined test or collection of tasks used to compare a system's performance under specified conditions.. The practical lesson is to measure recall policy as a first-class system metric because irrelevant memory consumes tokens and can send an agent in the wrong direction.
Local execution provided control over data, traces, compute and evaluation, but serial inference made experiments slow. The talk closes by placing these tradeoffs within the wider goal of sovereign AI.
Watch on YouTube
