in progress

eureka.ai

Eureka is an intelligent research platform that helps scientists discover connections, identify research gaps, and generate hypotheses using AI-powered knowledge graphs and advanced RAG (Retrieval-Augmented Generation).

---

Features

Core Capabilities

• Conversational Research (RAG): Natural language queries with AI-powered answers and citations using HuggingFace embeddings
• Knowledge Graph: Interactive visualization of research relationships and connections
• Autonomous Discovery: HuggingFace-powered AI agents identify research gaps, contradictions, and emerging trends
• Hypothesis Generation: HuggingFace models generate testable research hypotheses from identified gaps
• Pattern Recognition: Multi-agent analysis across domains and methodologies using HuggingFace models

Discovery Pipeline

1. Upload Papers → PDF processing and text extraction
2. RAG Chat → Query papers using semantic search (HuggingFace embeddings)
3. Discovery Analysis → HuggingFace models analyze documents for:
   • Research gaps identification
   • Hypothesis generation
   • Trend detection
   • Contradiction detection

Technical Highlights

• Modern Stack: React 18 + TypeScript + Vite + TailwindCSS
• Production Ready: Error boundaries, loading states, lazy loading, accessibility
• Scalable Backend: FastAPI with async/await, PostgreSQL, vector search
• Developer Experience: Hot module replacement, TypeScript strict mode, ESLint ready

---

Quick Start

1. Clone Repository

git clone <repository-url>
cd eureka.ai

2. Frontend Setup

# Install dependencies
npm install

# Create environment file
cp .env.example .env

# Start development server
npm run dev

The frontend will be available at http://localhost:3000

3. Backend Setup

# Navigate to backend directory
cd backend

# Create virtual environment
python -m venv venv

# Activate virtual environment
# Windows:
venv\Scripts\activate
# Mac/Linux:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Start backend server
python -m uvicorn app.main:app --reload --port 8000

The backend API will be available at http://localhost:8000

API Documentation: http://localhost:8000/docs

---

Development

Frontend Commands

# Development server with hot reload
npm run dev

# Build for production
npm run build

# Preview production build
npm run preview

# Type checking
npm run lint

Backend Commands

# Development server with auto-reload
cd backend
uvicorn app.main:app --reload --port 8000

# Production server
uvicorn app.main:app --host 0.0.0.0 --port 8000 --workers 4

# Run tests (if available)
pytest tests/

First-Time Setup

1. Install Python dependencies:

   cd backend
   pip install -r requirements.txt

2. HuggingFace models download automatically on first use:

   • Embedding model: sentence-transformers/all-MiniLM-L6-v2 (~90MB)
   • Discovery model: google/flan-t5-small (~300MB)

3. Optional: Configure models in backend/app/config.py:

   DISCOVERY_MODEL: str = "google/flan-t5-small"  # or flan-t5-base for better quality
   HF_USE_LOCAL_GENERATOR: bool = True

---

Configuration

Environment Variables

Create a .env file in the root directory:

# Backend API URL
VITE_API_BASE_URL=http://localhost:8000/api

# Optional: Debug mode
VITE_DEBUG=true

Backend Configuration

Backend configuration is managed in backend/app/config.py. Key settings:

• Database connection
• API keys (HuggingFace, etc.)
• Model configurations
• CORS origins

---

API Endpoints

Document Management

• POST /api/documents/upload - Upload a research paper (PDF)
• GET /api/documents - List all uploaded documents
• GET /api/documents/{id} - Get document details

Query Endpoints (RAG Chat)

• POST /api/queries/ask - Ask a question about uploaded documents

  {
    "question": "What are the main findings?",
    "document_id": 1,  // Optional: query specific document
    "top_k": 5
  }

• GET /api/queries/history - Get query history

Discovery Endpoints (HuggingFace-Powered)

• POST /api/discovery/analyze - Run full discovery analysis on all uploaded documents
  • Uses HuggingFace models to identify gaps, generate hypotheses, detect trends and contradictions
• GET /api/discovery/gaps - Get research gaps (from last analysis)
• GET /api/discovery/hypotheses - Get generated hypotheses
• GET /api/discovery/trends - Get trending topics
• GET /api/discovery/contradictions - Get detected contradictions

Document Endpoints

• POST /api/documents/upload - Upload a research paper
• GET /api/documents - List all documents
• GET /api/documents/{id} - Get document details

Full API Documentation: http://localhost:8000/docs

---