Currai
Observability & A/B Testing for LLM Apps
Start new thread

Currai - Find and fix failures in your AI agent conversations.

by•
Agents are taking on real economic work. As runs get longer, mistakes become expensive at best, catastrophic at worst. Today, teams find out from user complaints. Or never. Monitoring tools only catch what you told them to watch for: you set up the judges, you trace the issues yourself. But agents will always fail in ways you didn't predict. That's why we built Currai Currai understands what your agent is supposed to do, so it surfaces failures you didn’t know to look for, automatically.

Add a comment

Replies

Best
Maker
📌
Hey Product Hunt! I’m Yasser, founder of Currai. I’ve spent more than six years building software, and recently much of that work has involved LLM-powered products. One problem kept coming up: traditional logs could show that an AI agent completed a conversation, but not whether it understood the user, followed policy, used its tools correctly, or produced the right outcome. That’s why I built Currai. Currai analyzes real chat and voice-agent conversations alongside model responses and tool calls. It helps teams: - Discover recurring user needs and emerging intents - Detect silent failures, violations, and conversation errors - Trace problems back to the exact model response or tool call - Understand why an agent failed using supporting evidence - Receive alerts in Slack when important issues appear - Identify improvements based on production conversations You can connect Currai using its SDK or existing OpenTelemetry instrumentation, then inspect conversations and traces from one dashboard. Currai is free to try at . I’d love to hear how you currently monitor your AI agents, what remains difficult to diagnose, and what you’d like Currai to support next.