From Real-Time ML to Agents with Hopsworks - Feature Store Summit 2025

From Real-Time ML to Agents with Hopsworks Hopsworks supports building batch, real-time, and agentic AI system using a unified architecture around feature pipelines, training pipelines, and inference pipelines. In this talk, we walk through the journey of developing batch and real-time ML systems to agentic AI systems with this unified FTI pipeline architecture. We will look in particular at how we connect application state to agents using "application context protocol". That is, we will see how entity IDs in applications can be used to help agents reliably retrieve application state as context. About the Speaker: Jim Dowling is CEO of Hopsworks and a former Associate Professor at KTH Royal Institute of Technology. He is a co-inventor of the open-source Hopsworks Feature Store platform, the organizer of featurestore.org, and has written a book for O'Reilly on building AI systems 00:00 Welcome 01:31 2025 state: consolidation and agents 06:04 Book and giveaway 10:16 Talk kickoff: real-time ML to agents 13:03 Lakehouse-first vs real-time-first 16:28 Data transformation taxonomy: shift-left/right 21:05 On-demand aggregations and pushdown 24:45 When to shift-left to streaming 28:40 Data modeling for feature stores 34:11 Implementing nested joins 38:33 From real-time ML to agents 40:52 App context via feature store 43:44 Traces, evaluations, and improvements 44:49 Hopsworks roadmap and closing