Yosr Kharrat – software engineering student

About Me

A third year software engineering student at the national institute of applied sciences and technology in Tunisia, passionate about building intelligent systems that bridge research and production. Worked on several proj…

Experience

AI/ML Intern

VistaDeep

Built an end-to-end RAG architecture combining document embeddings, vector search (FAISS), and LLM-based response synthesis using Hugging Face Transformers and LangChain, Developed an AI-driven quantitative trading platform combining XGBoost forecasting and PPO-based portfolio optimization, Scraped and processed 10,000+ research papers via arXiv API, Developed an unsupervised anomaly detection system trained on 1M+ HDFS log sequences, Analyzed 50 years of environmental data (148+ indicators) including CO2, CH4, N2O and PM2.5

Skills

C++ Language (C++) Java PHP: Hypertext Preprocessor (PHP) Python Structured Query Language (SQL) JavaScript NumPy CI/CD Docker PyTorch Hugging Face Transformers XGBoost Stable-Baselines3 LangChain Google Cloud Run GitHub Actions REST APIs pandas Matplotlib Seaborn Plotly Reinforcement Learning (PPO) Feature Engineering Time Series Forecasting Model Explainability RAG Architectures Prompt Engineering Vector Databases Embedding Models Reinforcement Learning PPO Production ML Systems