Experience
L’ETNIC
- Position: Full Stack Developer Intern
- Location: Brussels, Belgium
- Period: 02/2024 - 05/2024
Description
During my internship at L’ETNIC, the IT partner of Fédération Wallonie Bruxelles, I worked at their Innovation Center. My task was to develop an internal web platform focused on document management, with a unique feature: integrating a chatbot for discussions with documents.
The main goal was to streamline document-related tasks and facilitate discussions within the organization. Using a Retrieval Augmented Generation (RAG) system, the chatbot allowed users to interactively discuss with documents, providing a more intuitive way to access information.
Throughout the internship, I collaborated closely with a small team of developers to design and implement this project from scratch. We focused on creating a user-friendly interface that seamlessly integrated the chatbot functionality into the document management system.
This experience allowed me to enhance my technical skills and gain valuable insight into integrating innovative solutions into real-world projects. It was also my first interaction with generative artificial intelligence and its use cases in practical applications.
Achievements
- Designed and developed an internal web platform for discussion and document management.
- Developed the application’s REST API using FastAPI, integrated it with a MongoDB persistence system, and implemented integration tests with pytest.
- Implemented an asynchronous task execution system (Task Queue) using Celery and Redis, enabling the management of requests and long-running background processes.
- Developed various reusable components in Angular 17, including advanced form templates, navigation components, and data management services.
- Implemented a conversational chatbot using a Retrieval-Augmented Generation (RAG) architecture, leveraging generative AI and vector search to provide users with contextualized responses based on documents.
Tech Stack
- Frontend: Angular, TypeScript, Angular Material, HTML, SCSS
- Backend: FastAPI, Celery, Python, LangChain
- Database: MongoDB (Atlas), Redis
- Other: Docker, Azure, Git, Mistral AI