Skip to content

shao3d/Experts_panel

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

409 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Experts Panel

CI/CD License Python React Tailwind

Intelligent system for analyzing expert Telegram channels and Reddit communities using Google Gemini AI

Experts Panel is a powerful tool for semantic search and analysis of content from expert Telegram channels and Reddit communities. The system uses an advanced 10-phase Map-Resolve-Reduce pipeline architecture with Google Gemini AI (2.5 Flash Lite & 3 Flash Preview) to provide accurate and contextually relevant answers.

🏗️ System Architecture

The system uses an advanced ten-phase pipeline to provide accurate and contextually relevant answers. The architecture includes cost-optimized Gemini-only strategy, differential processing for posts based on relevance, and parallel pipelines for content and comment analysis.

Models Strategy (Gemini Only)

  • Map Phase: Gemini 2.5 Flash Lite (Speed & Instruction Following)
  • Synthesis/Reduce: Gemini 3 Flash Preview (Reasoning)
  • Validation/Analysis: Gemini 2.0 Flash (Speed)

✨ Key Features

  • 🎨 Modern UI: Clean, responsive interface built with React, Tailwind CSS, and a collapsible Sidebar.
  • 🧠 10-phase Map-Resolve-Reduce Architecture: Advanced pipeline with differential HIGH/MEDIUM posts processing.
  • 🎯 Cost-Optimized Gemini Strategy: Google AI Studio with Tier 1 account (high rate limits).
  • 🔍 Smart Semantic Search: Finds relevant posts by meaning using Hybrid Retrieval (Vector KNN + FTS5 + RRF).
  • 📊 Medium Posts Reranking: Gemini-based scoring system with threshold ≥0.7.
  • 💬 Comment Groups & Synthesis: Gemini pipeline for comment drift analysis.
  • ⚡ Real-time: Server-Sent Events (SSE) for instant progress updates.
  • 👥 Multi-expert Support: "Select All" groups, smart avatar initials, and complete data isolation.
  • 🕒 Smart Filters: "Embs&Keys" (Hybrid Search), "Recent Only" (3 months), and "Reddit Search" toggles directly in the Sidebar.
  • 👽 Reddit Integration: Parallel analysis via Sidecar Proxy (experts-reddit-proxy).
  • 🚀 Reddit-Only Mode: Bypass expert analysis for broad community searches.

🚀 Quick Start

Prerequisites

  • Python 3.11+
  • Node.js 18+
  • Google AI Studio API key(s)

For a guided setup experience, execute the quickstart.sh script located in the project root.

./quickstart.sh

📚 Documentation

🤝 Contributing

  1. Fork the repository
  2. Create your feature branch (git checkout -b feature/AmazingFeature)
  3. Commit your changes (git commit -m 'Add some AmazingFeature')
  4. Push to the branch (git push origin feature/AmazingFeature)
  5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

About

Experts Panel - Multi-expert query processing system with AI integration

Topics

Resources

License

MIT, OFL-1.1 licenses found

Licenses found

MIT
LICENSE
OFL-1.1
OFL.txt

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors