rb.

My Projects

A showcase of my work in AI, MLOps, and software development, ranging from custom frameworks to full-stack applications.

RAGLight Framework

RAGLight is a lightweight and modular Python library for implementing Retrieval-Augmented Generation (RAG). It enhances the capabilities of Large Language Models (LLMs) by combining document retrieval with natural language inference.

Current Features :

- LLM Agnostic
- RAG Pipeline
- RAT Pipeline
- Agentic RAG Pipeline

Python Framework LLM RAG AI Agents Langchain LangGraph

Coding Assistant

AI Coding Assistant powered by MCP server : it reads your code locally to provide truly context-aware help, explanations, and suggestions directly related to your files. Code smarter, locally or using external API.

Python Streamlit MCP LLM FastAPI

RAG Scientific Papers

End-to-end automated system fetching daily arXiv papers, processing PDFs, indexing via RAG (ChromaDB), storing files (MinIO), and providing a chat interface (React/FastAPI). Orchestrated with Prefect & Docker.

Python Prefect RAG Docker FastAPI MinIO ChromaDB React

Pneumonia Detection App

Full-stack web application demonstrating the deployment of a fine-tuned Vision Transformer (ViT) model for pneumonia detection from chest X-rays. Built with React, FastAPI, Nginx, and Docker.

Computer Vision Fine-Tuning ViT Model MLOps Docker FastAPI React Nginx

Photo Gallery Viewer

An offline-capable desktop application leveraging multi-modal AI (CLIP) for semantic image search. Uses image embeddings stored in ElasticSearch to allow filtering photos via natural language queries.

Python CLIP Model Embeddings ElasticSearch Tkinter Computer Vision