Sumit Agarwal Email & Phone Number

Moto : Make AI run faster, faster and faster!Interest: GPU Compute Programming, AI accelerator, NLP (Transformer), Distributed Cloud Computing Platform, AI Infrastructure.I work on building GPU accelerated high-performance C++ low-level public ML platform (DirectML) and GPU-agonistic runtime of high-level ML frameworks (ONNX Runtime) to expedite ML inference. I have recently completed a Master's in Computer Science from Stony Brook University, New York. In the past, I have worked at Uber and Goldman Sachs. In Uber, I got an opportunity to design and build a configuration-driven gRPC-based UX experimentation service platform in GO, to allow runtime customization. In Goldman Sachs I was under Core Engineering Division to proactively identify Data Quality issues in financial products, expose the ML model as RESTful APIs, and efficiently reconcile two datasets, by building and launching software that harnessed Reinforcement Learning, Web-services, and Data Structure & Algorithm.The professional experience gave me the opportunity to showcase an ability to comprehend business problems and recommend an efficient technical solution.