Agent Studio
Know what the AI is doing—and make it better over time.
Observability for AI analytics
See how the agent is used across your team—conversations, tokens, cache efficiency, and active users—all in one dashboard.
- Time-windowed usage overview with period selection.
- Search across messages with filters by scope, role, user, and conversation.
- Covers notebooks and semantic layers.
CONVERSATIONS
20
TOTAL TOKENS
1.4M
ACTIVE USERS
3
CACHE HIT RATE
66%
Improve reliability
Find patterns in where things go wrong—tool calls, context gaps, ambiguous prompts—and fix them systematically instead of one-off.
- Identify which workflows need guardrails or templates.
- Reduce support burden by addressing repeat failures.
- Track improvements over time with usage trends.
Improve outcomes
Spot high-impact questions, understand which semantic layers drive the best results, and make agent behavior measurable.
- Elevate frequent questions into reusable workflows.
- Compare agent performance across different contexts.
- Turn opaque AI into something teams can reason about.
AI that gets better with use
If AI is part of production analysis, teams need more than “it answered a question.” They need to see what happened, identify failure modes, and improve prompts, context, and workflows over time.