Data Scientist
- Développement informatique R&D
Working on recommender systems, click-through rate and real-time bidding.
I've spent my last year project as an apprenticeship working on a bid optimizer for real-time bidding system using deep learning. In order to apply a reinforcement learning algorithm, I've built a framework in Python to train and evaluate my model. The project was quite similar (in my opinion) as a research project (study the state-of-the-art, coding from scratch, try new ideas, comparison with SoTA models).
Softwares/languages/librairies used : Python (pandas, numpy, scikit-learn, lightgbm, keras, sphinx), git
I've also worked on a SaaS project where I was in charge of building the data pipeline and one of the recommender system.
Softwares/languages/librairies used : Kafka, Storm, Cassandra, Java, Python (pandas, PyTorch, scikit-learn, numpy)
Keywords : Machine Learning, Deep Learning, Reinforcement Learning, Auto-Encoders, Q-learning, Apache Storm, Apache Kafka