Chaima E.

Data Scientist @Air Liquide

Paris, FR

Development of a RAG model based on technical documentation (markdown/html/pdf) with LlamaIndex to create a search engine using LLMs (Claude3)- Integrating the chatbot into Google Chat using Google Cloud Plateform (GCP)- Data collection from various sources through APIs, cleaning, statistical computing, and creation of Power BI dashboards to exploit results for different clients- Classification of documents into confidential and non-confidential based on their title and metadata - Creation of a tool for WordCloud generation using Spacy- Clustering IT incidents based on their description using Bertopic, LLMs and other NLP techniques to anticipate potential incidents-Development of NLP model for personal and confidential data detection in the text; exploring multiple NLP models: Tf-idf, BERT, Claude3, and open source models from HuggingFace- Creating Streamlit webapps for the demo

Ecole Supérieure Privée d'Ingénierie et de Technologies - ESPRIT

Preparatory school

National Engineering School of Tunis (ENIT), University of Tunis El Manar

Engineer's degree, Advanced Techniques

Institut Polytechnique de Paris

Master's degree, Data science

ENSTA

Master of Engineering - MEng, Applied Mathematics