I'm currently the Technical Program Manager (TPM) for Anyscale, a platform that specializes in distributed machine learning using Ray. The pitch is simple: Ray is the best place for distributed machine learning and Ray works best on Anyscale :).Previously I worked on Michelangelo, Uber's AI/ML platform for deep learning, development experience, and feature embeddings. The platform powered the ML solving a variety of Uber's challenges ranging from safety, to ETA predictions, to Eats ranking and much, much more (DM me if you ever want to nerd out ^3^). I'm familiar with the full range of the ML pipeline from storage, data, feature prep, transform, training, evaluation, and serving (albeit my bread and butter is still in the infrastructure powering all of the above at the end of the day).I also worked with the Hololens, Azure, and Microsoft Research teams in NLP and CV ML (eg: object detection, classification, tracking, search, semantics, and correlations to name a few), specifically around generating synthetic datasets leveraging the power of distributed computing on the Cloud.About 75% of my time working on the above have been on the infra side of things; although I've driven ML performance programs working directly with researchers, data scientists, and machine learning engineers as well! I also spent years working on Azure Government clouds, which are physical, logical, and network isolated instances of Azure for the US government at various levels of classification (from Civilian all the way to S/TS).In my college years I "founded" two startups (and am always tinkering on something. My (current) passion lies in AI and ML ^_^.