I have worked on numerous projects in machine learning and deep learning. Here are some of the notable open source contributions:

• Graphium: A graph machine learning library optimized for molecular tasks, emphasizing multi-tasking, multi-level representation learning, and extensive models/features for diverse molecular assignments.
• surveyweathertool-dssg23-jumpstart: A library designed for spatio-temporal analysis of climate impacts and poverty, integrating survey and weather data.
• Prepared tutorials for the Deep Learning Indaba conference: Tutorial 2022 and Tutorial 2023.
• Prepared tutorials for the Neuromatch Academy summer school: Tutorial 1 and Tutorial 2.
• SAINT Implementation: Improved Neural Networks for Tabular Data via Row Attention and Contrastive Pretraining.
• Kernel methods for DNA sequence region prediction: Predicts if a DNA sequence region binds to a specific transcription factor.
• Face recognition system: A fully functional system, adopted by various industries.
• Very deep CNNs for raw waveforms: Implementation of very deep convolutional neural networks for raw waveforms.

For a complete list, visit my GitHub profile.