Documentation
Technical guides for integrating HumanJudge with your AI application. Add human evaluation from verified experts to your Langfuse monitoring stack.
What is HumanJudge?
HumanJudge adds a human evaluation layer to your AI monitoring stack. Verified domain experts (content creators, professionals) evaluate your AI outputs, with scores syncing to your Langfuse dashboard and appearing on public evaluation pages.
How It Works
┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐
│ Your AI App │────▶│ GrandJury SDK │────▶│ Langfuse │
│ (Frontend) │ │ (grandjury.js) │ │ Dashboard │
└─────────────────┘ └─────────────────┘ └─────────────────┘
│ │ │
│ ▼ │
│ ┌─────────────────┐ │
└─────────────▶│ Chrome Extension│◀─────────────┘
│ (Reviewers) │
└─────────────────┘
│
▼
┌─────────────────┐
│ Public Verdicts │
│ + Scores │
└─────────────────┘Quick Links
Key Features
Langfuse Integration
Human evaluation scores sync directly to your Langfuse dashboard alongside automated metrics.
Session Discovery
SDK automatically detects your existing session patterns and maps them for evaluation tracking.
Inference Tracking
Track individual AI queries with unique inference IDs for granular evaluation.
Chrome Extension Compatible
SDK exposes session data for seamless Chrome extension detection by reviewers.
Quick Start
Add GrandJury to your app in under 5 minutes:
<!-- Add to your HTML -->
<script src="https://grandjury.xyz/sdk/grandjury.min.js"></script>
<script>
GrandJury.init({
projectId: 'proj_your_project_id'
});
</script>See the Integration Guide for complete setup instructions including Langfuse configuration.
Need Help?
Questions about integration? Email support@humanjudge.com