Organizers

Emily Webber, egwebber@amazon.com, is a Principal Machine Learning Specialist Solutions Architect at Amazon Web Services. She has assisted hundreds of customers on their journey to ML in the cloud, and in particular specializes in distributed training for large language models. Emily has organized 50+ hands-on workshops for hundreds of participants. She has also co-organized and served as a chair for internal-to-Amazon research workshops on systems optimizations for ML. She has mentored 60+ Machine Learning Solution Architects, authored countless feature designs for SageMaker and AWS, and guides the SageMaker product and engineering teams on best practices for customers. Emily is widely known in the AWS community for a 2018 YouTube video series featuring Amazon SageMaker with 160,000 views, along with a Keynote at O’Reilly AI London 2019 on a novel reinforcement learning approach she developed for public policy.

Fouad Bousetouane, fouad.bousetouane@grainger.com, is a Senior Principal Applied Machine Learning Scientist at W.W. Grainger, working on building cutting-edge AI-powered capabilities to solve a wide spectrum of scientific and business problems. His research revolves around data-driven decision-making, Computer Vision, Hyperspectral Imaging, Deep Learning, Multimodal Machine Learning, and Natural Language Processing. He authored 40+ research papers and book chapters, filed 10+ patents, and owns multiple high-performance Edge-AI systems trade secrets. Fouad received multiple awards for his contribution in building best-in-class vision-AI solutions, including Timmy-Award Best-tech-manager in Chicagoland 2020, Best innovation of the year from AIJRF-Dubai, and featured on MIT Technology Review Arabia as a Top-30 AI scientist. Fouad received his Doctorate in Artificial Intelligence and Pattern Recognition from the University of Badji Mokhtar-Annaba in collaboration with the LISIC research lab and CNRS-Lille, France. He holds a master's degree in Artificial intelligence and a bachelor's degree in Computer Science and Mathematics from Badji Mokhtar University. He also received a Doctorate-equivalent degree (Ph.D.) accredited by the University of Nevada, Las Vegas in 2015. He worked as a postdoctoral researcher at the Real-time Intelligent System Laboratory, University of Nevada, Las Vegas, and co-supervised multiple Ph.D. and master's in science theses. He has also served as a reviewer for top-ranked journals and conferences, including IET-Image Processing Journal, IEEE Transaction Journal on Intelligent Transportation Systems, IEEE-IROS2012, and IEEE-ITSC-2015.

Can Karakus, cakarak@amazon.com, is a Senior Applied Scientist at Amazon, working on large-scale distributed training of deep learning models. He has received his B.S. from Bilkent University in 2011, M.S from UCLA in 2013, and PhD from UCLA in 2018, all in electrical engineering. His research spans a wide range of areas from theory and systems domains, including machine learning, distributed and high-performance computing, optimization, information theory, and wireless networks, and he has published 20+ research papers and 3 patents across these fields. At Amazon, he has envisioned, designed, and contributed to the development of the SageMaker distributed machine learning services. He is a recipient of UCLA Graduate Division Fellowship, UCLA EE Department Fellowship, and Qualcomm Roberto Padovani Award. He has also served in the Technical Steering Committee of open-source distributed training library Horovod, and in Technical Program Committees of NeurIPS and ICML.

Arjun Balasubramanian, balarjun@amazon.com, is a Senior Software Engineer at Amazon Web Services focused on building high-performance, hardware accelerated collective communication algorithms for distributed deep learning. He is broadly interested in systems for large-scale machine learning, with prior graduate research tackling a broad range of problems including fair and efficient GPU cluster scheduling for deep learning workloads (Themis; NSDI’2020), network communication scheduling in multi-tenant GPU clusters (MLfabric; SoCC’20), low latency inference for large models (Freeze Inference; HotCloud’19, Gati), and serverless control plane design for low latency workloads such as inference (Atoll; SoCC’21). Arjun serves as an expert reviewer in topics such as distributed training, inference, and RDMA at peer-reviewed conferences and journals,

Corey Barrett, corbarre@amazon.com, is a senior data scientist in Amazon’s Machine Learning Solutions Lab assisting customers with building machine learning models on AWS, with a specialization in distributed training and computer vision. Corey’s research interests are in Distributed Training, Federated Learning, and Adversarial Learning, with a history of applying those to Computer Vision use-cases. Corey has experience leading successful workshops in privacy and federated learning as well as a tutorial on distributed training for MXNet at internal Amazon Conferences. Corey has delivered multiple hands-on workshops to customers demonstrating model training including distributed training, delivering content to over 150+ participants. Corey has developed and delivered 20+ machine learning solutions for customers which addressed novel customer problems in unique ways.

Anish Mohan, nishm@nvidia.com, is a Principal Deep Learning Architect at NVIDIA. He is responsible for helping effective usage of GPU for running Deep Learning workloads on Cloud. He has helped many NVIDIA & Cloud customers todo large scale training and inference on GPUs. His work on training large scale Transformer and MaskRCNN models on 2K+ GPUs, which has been featured in keynote in Amazon re:Invent. He has worked with various AI/NLP start-ups for faster training of NLP models. For example this work with Deepset.ai improved the training time for BERT by 4x. He has organized more than 25 hands-on workshops on deep learning training and inference with GPUs. As a member on the Advisory Board for University of Washington, Professional College of Education's Data Science And Machine Learning Program, he has helped to design course offerings to further the reach of AI & ML. He is also a creator of the Coursera Course on Machine Learning. Anish is a Senior IEEE Member and the General Secretary for the Seattle Chapter for IEEE's Computer Society. He is also a reviewer for IEEE Transactions on Geoscience and Remote Sensing, Journal Of Electronic Imaging, IEEE Open Access Journal and PeerJ Journal for Computer Science. Before NVIDIA, Anish was at Microsoft's Artificial and Research group helping to set-up Deep Learning/Machine Learning Services on Azure. At Microsoft, he was awarded a patent for his work to use Image processing techniques for rendering verification.