I'm an applied research scientist working primarily on Bayesian active learning, uncertainty estimation, and explainability. I want to make AI works for everyone and I believe that it starts with better practices around dataset creation and model inference.
I'm slowly learning frontend development with React 🚀 and I have strong skills as a backend developer (FastAPI, Dask, HuggingFace, Pytorch).
I'm a strong believer in opensource, I was chairman at Keras before co-founding Baal, an active learning library. Recently, I worked on opensourcing Azimuth, an error analysis tool for text classification.
I'm looking for new projects to contribute to, please contact me if you need some help! I am learning more about the HuggingFace 🤗 and Pytorch Lightning ecosystems which I find fantastic!
Ask me about anything related to (fair) deep learning and open-source!
My socials: