- Visualize agent trajectories: See what your agent actually did, every tool call, error, and recovery, in seconds
- Detect issues automatically: default signals catch common failure modes like forgetting, user frustration, and task failures
- Search across millions of interactions: Use Deep Search to find specific issues in your production data with natural language
- Track custom signals: Define any signal you care about (eg. syntax errors, aesthetic complaints, agent stuck in a loop) and track it at scale
- A/B test your agents: Run experiments to validate that your fixes actually worked
Integrate Raindrop
TypeScript
Full-featured SDK with tracing support for Node.js and edge runtimes.
Python
Native Python SDK for FastAPI, Django, and other Python frameworks.
HTTP API
RESTful API for any language or platform.
Vercel AI SDK
Automatic tracking for the Vercel AI SDK with zero configuration.
Core Features
Trajectories
Visualize and search agent traces. See every tool call, spot errors instantly, and understand what actually happened.
Signals
Ground truth indicators for agent performance. Track wins, failures, and custom behaviors across all your interactions.
Deep Search
Find issues in your production data using natural language. Like deep research, but for your agent logs.
Experiments
A/B test your agents to validate changes. Compare models, prompts, and configurations.
Alerts
Get notified via Slack when issues spike. Daily summaries and custom alert thresholds.