High-performance RAG (Retrieval-Augmented Generation) engine built with Go. This service handles document ingestion, text extraction, and vector-based search.
- URL & PDF Ingestion: Automatic text extraction from web pages and documents.
- Vector Embeddings: Generates semantic vectors using OpenAI's
text-embedding-3-smallmodel. - Bilingual Brain: Adaptive AI response system supporting English and French.
- Secure Authentication: JWT-based user context management.
- Real-time Processing: Fast response times optimized for RAG workflows.
- Language: Go 1.22+
- AI: OpenAI API (GPT-4o mini)
- Database: PostgreSQL with
pgvector - Tools: Go-OpenAI SDK
- Define your environment variables in
.env:OPENAI_API_KEY=your_key DATABASE_URL=postgres://... JWT_SECRET=your_secret
go run cmd/main.go