Graph Neural Networks (GNNs) have emerged as the most effective approach for machine learning on graphs. Graphs are useful for describing most real-world applications (including financial graphs, social networks, e-commerce transactions, online marketplaces, and more), where phenomena can be broken down into entities, their relationships, and their dynamic interactions.
However, GNNs at scale present a number of technical and operational challenges around continuous data loading and ingestion, mini-batch sampling, and concurrent training across multiple machines. In addition, executing predictions within a reasonable timeframe requires significant infrastructure designed around a microservices architecture.
At Kumo, we solved these challenges and evolved our platform to handle enterprise applications. In this webinar, we’ll perform a comprehensive deep dive into our approach for addressing these challenges. We’ll show how we built an efficient system using a microservices architecture that handles scaling the data, compute, and graph engines to deliver graph learning at scale.
Founding Engineer
Kumo AI
Subramanya Dulloor is a computer scientist and engineer with extensive experience in distributed systems and systems for ML. As a founding engineer at Kumo.ai, he is focused on making predictive analytics on data warehouses plug-and-play easy. Before joining Kumo, Dulloor worked on distributed systems and database internals at various startups. He also spent several years at Intel Labs conducting research in memory management in operating systems and VMMs, as well as in distributed high-performance systems for machine learning and computer vision.
Dulloor holds a PhD in computer science from Georgia Tech and has over a dozen publications in top-tier conferences and patents to his name.
Founding Engineer
Kumo AI
Dong Wang is a seasoned tech lead with extensive experience in machine learning and infrastructure. He is currently leading the ML platform work at Kumo.ai, a company that provides a no-code ML platform and predictive query to convert business problems into machine learning problems.
Prior to joining Kumo, Dong held leadership roles in several high-profile companies, including Coupang, Wish, and Pinterest, where he led the search and recommendations teams. He was also served as the leader of the machine-learning platform at Nuro before. Dong's work centers on making machine learning models more scalable, efficient, and generating better output.
Platform Engineer
Kumo AI
Min Shen is an experienced tech lead with a track record of leading multiple platform engineer at Kumo.ai. Prior, Min spend 8 years at LinkedIn, focused on building and scaling LinkedIn's general purpose batch compute engine based on Apache Spark to power use cases ranging from data analytics, data engineering, ML feature engineering, and model training.
Min holds a PhD in computer science from University of Illinois, Chicago, where he was performing research in computation and software design.