TestedliveAgentic Memory System
Recall Probe
Ask a servicing agent's memory the same question two ways, and watch one of them get it wrong.
DDOVDavid and Oliver
Eval labelTask eval
- Cases
- 8
- Passing
- 7
- Failing
- 1
- Built
- 21 Aug 2026
- Quote until
- Verified until 19 Nov 2026
Try it
The working version.
Mocked for this demo, with the numbers taken from the run that produced it.
Corpus
40 sessions · 62.4k tokens · 11 weeks · one insurance servicing casePick a question to run it against both memory designs.