Guillaume Tauzin Email & Phone Number

I am passionate about the science behind machine learning and the development of machine learning pacakages. I particularly appreciate the scikit-learn ecosystem to which I contributed different libraries for medium-scale deep learning or time-series forecasting (closed-source at INAIT). During my postdoc at EPFL, I created giotto-tda, an open-source Topological Data Analysis library for feature engineering and unsupervised learning. My interests span statistics and classical machine learning. In particular, I have had a strong focus om time-series-related tasks, especially, time series forecasting and time-series classification. My strong curioisity drives me to acquire a deep understanding of the concept I manipulate. My engineering background gave me the tools required to set up ML platform, ETL pipelines for data manipulation and develop systematic processes for training, tuning and evaluating models, as well as deploying model in production while monitoring them.My background is diverse and interdisciplinary, spanning mathematics, theoretical physics, high-performance computing, and sustainability science. I am fluent in Python and C/C++, and proficient in various ML and data handling/visualization libraries, such as scikit-learn, pytorch, pytorch-lightning, optuna, mlflow. I am passionate about learning and discussing new topics and ideas in ML and related fields, and I welcome any opportunities to collaborate or share my expertise especially if it is applied to projects linked to transition to sustainable development and climate change.

Listed skills include Lattice Boltzmann Methods, Cfd, Fluid Dynamics, Turbulence Modeling, and 42 others.