Agnost AI: Analyzing Agent Conversation Feedback
laalshaitaan · hn · 2026-07-15
Agnost AI has launched on HN, positioning itself as "product analytics for conversations" tailored for chat and voice agent teams.
It ingests production conversations to automatically detect behavioral failures: users repeating themselves, cursing at the agent, correcting it, asking for missing features, or leaving despite getting an answer. The team integrates messages via SDK or OTel, combining metadata like account, plan, source, and organization to cluster conversations into product intents. It identifies feature requests and bugs by default, while other categories evolve dynamically based on customer data.
The founders emphasize that this product differs from observability or evals: the former monitors technical events, and the latter validates known cases. Instead, they focus on "what users actually want, where they get stuck, and what new evals should be added."
More from coding & agent
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22
- Agent search bottlenecks are now about variance, not raw latency — rohanpaul_ai · 2026-07-22
- LangSmith adds tracing for Pipecat, LiveKit, OpenAI Realtime, and Gemini Live — LangChain · 2026-07-22
- An MCP server signs every AI agent tool call into a verifiable Merkle chain — Funky_Chicken_22 · 2026-07-22
- Annotated transcript of a Claude Code team interview is now available — trq212 · 2026-07-22
- Claude Code skill uses 10 Markdown rules to make outputs ADHD-friendly — alex_verem · 2026-07-22