Most voice agents forget the moment the call ends. Accellio gets better because of it. It briefs a seller before a CFO meeting from public signals, then reads the transcript itself - nobody types notes. It grades that briefing, rewrites its own
briefing policy, and regenerates the briefing. Between the meeting ending and the better briefing appearing, human inputs: zero.
The hard part is proving "it improved" isn't self-graded. Senso verifies each claim against its own ground truth and contradicted three of ours before our evaluator ran. Evaluation is deterministic checks first, Senso second, a model
rubric last.
The learning is mechanical, not cosmetic. Patches change evidence-reliability weights feeding Actian's vector ranking, so the next briefing retrieves different sources. Pioneer runs reasoning and embeddings.
It also knows what it isn't entitled to decide. Down-weighting an evidence source across every account, on one meeting, is escalated to a human through Band — and the loop carries on without waiting. A reviewer agent we published to Guild, on
their runtime, gives an independent second opinion. It usually says no.
Five sponsor tools serve, each with a working fallback.
