RAG Assistant Docker

Applied AI prototype of an educational assistant based on retrieval-augmented generation.

The project is designed to index local course materials, retrieve relevant fragments, and generate answers with source-aware context. It uses a backend service, a simple frontend, and Docker-based execution.

Main idea

The system provides an educational question-answering workflow over local documents.

It was developed as a practical RAG prototype for working with structured and unstructured learning materials.

Main features

• Local document indexing
• Vector search over educational materials
• Retrieval-augmented question answering
• Source-aware answer generation
• Backend service for RAG interaction
• Simple web frontend
• Docker-based deployment
• CLI mode for local testing

Technical stack

• Python
• RAG
• Vector search
• Local LLM integration
• Docker
• HTML / JavaScript / CSS frontend

Repository structure

• backend/ — backend logic, indexing, and LLM utilities
• frontend/ — simple browser interface
• cli_mode.py — local command-line mode
• Dockerfile — container setup
• docker-compose.yml — container orchestration
• requirements.txt — Python dependencies

Data note

Course materials, PDFs, DOCX files, spreadsheets, vector indexes, model files, local caches, generated outputs, and private environment files are not included.

This repository contains code only.