I create, and lead teams which create, Machine Learning algorithms to solve problems in computer vision, biomedical image understanding, NLP, and EHR (electronic health record) understanding. I'm always happy to enter into new domains, particularly where the data are well-structured. Deep neural networks (CNNs, Autoencoders, Transformers etc.) are my tools of choice for image and signal data, whereas for tabular data I typically stick to 'traditional' ML methods like logistic regressions and random forests , augmented with techniques like bootstrapping and multiple imputation. My DL teams work closely with product and engineering teams that build out the full infrastructure around these models and deploy them to production.I also have extensive experience in software engineering (esp. C++) and have loved using it to create products and applications in ML, signal processing, data analysis and visualization. Before entering industry I was an experimental and computational neuroscientist, where I acquired good general conceptual tools that I still use to help attack my current research problems.In my spare time, I develop my theoretical background (Probability Theory, Optimization, Functional Analysis, Optimal Transport Theory...). I have built and trained donkey cars (www.donkeycar.com). I continually seek to grow myself professionally and play an active part of the thriving tech community in the Triangle, both as a participant and occasional organizer of ML-related Meetups.Banner image credits: 1. Matt Phillips2. Diagnostic and Interventional Cardiology, https://www.dicardiology.com/article/dose-lowering-practices-cardiac-ct-angiography3. Google Earth3. Matt Phillips4. Kaggle text denoising, https://www.kaggle.com/c/denoising-dirty-documents/data