Agnost AI Launches Product Analytics for Chat and Voice Agents

Agnost AI is a new product analytics platform designed for teams building chat and voice agents. The service reads production conversations and automatically clusters them into product‑specific intents, surfacing behavioral failures such as rage‑prompting, repeated re‑phrasing, and missing features. Agnost ingests conversation data via SDKs or OpenTelemetry, optionally tagging each event with account, plan, source, or organization metadata. The platform uses a hybrid clustering pipeline that first segments conversations by cosine drift, then compresses the candidate space with BIRCH, and finally applies HDBSCAN‑style clustering to discover new patterns.

Agnost AI offers a free Starter tier, a $499/month Pro tier, and an Enterprise tier for high volume and security needs, and it is SOC 2 Type 1 compliant with Type 2 in progress. The SDKs are available on PyPI and npm, and the service stores data only for the customer that ingests it. By automatically generating feature‑request clusters, the platform helps product teams prioritize new capabilities without manual labeling. Agnost AI is positioned as a discovery tool rather than an observability or evaluation framework.