Lucas Collemare

Below are a few examples of projects I have completed: • Road Traffic Simulation (Data Engineering) Design and implementation of a real-time data pipeline using Kafka, Scala, MinIO (S3), and PostgreSQL to detect priority vehicles and centralize data. • Legislative Text Generation (NLP) Development of a classifier trained with PyTorch to generate and categorize excerpts from legal texts. • Recommendation Systems (Machine Learning) Development of a complete recommendation pipeline combining collaborative filtering (ALS) and content-based methods (TF-IDF) on the KuaiRec dataset. Feature extraction using Pandas, model implementation in Python, and performance evaluation (Precision@K, NDCG@K). • Atari DQN (Deep Reinforcement Learning) Full implementation of a Deep Q-Network agent in PyTorch for the Breakout (Atari) environment. Design of a convolutional neural network, performance analysis through reward tracking, and generation of evaluation videos of the trained model.