FSS/25

[ AGENDA // 15 ]

Oct 14, 06:30 AM UTC · 10 min · Kickoff
› Kick - Off
Jim Dowling kicks off the event with insights on how feature stores are transforming Real-Time AI.
Oct 14, 06:40 AM UTC · 35 min · Presentation
› From Real-Time ML to Agents with Hopsworks
Learn to build batch, real-time, and agentic AI systems using a unified FTI pipeline architecture.
Oct 14, 07:15 AM UTC · 25 min · Presentation
› Lyft’s Feature Store: Architecture, Optimization, and Evolution.
Learn about Lyft’s Feature Store: its architecture, key use cases, performance, and evolution over 5+ years.
Oct 14, 07:40 AM UTC · 25 min · Presentation
› Powering Real-Time AI at Pinterest: Feature Management and Serving at Scale with Galaxy and Scorpion.
Learn about Pinterest’s Galaxy Platform: architecture, real-time feature serving, and lessons at scale.
Oct 14, 08:05 AM UTC · 25 min · Presentation
› Vector Store: Uber’s Embedding Platform
Learn how Uber’s Vector Store powers embeddings at scale for GenAI, RAG, semantic search, and predictive models.
Oct 14, 08:40 AM UTC · 25 min · Presentation
› From EC2 to K8s: Zalando's Journey to Large-Scale, Real-Time Feature Serving.
Learn about How Zalando Runs Real-Time AI at Scale.
Oct 14, 09:05 AM UTC · 25 min · Presentation
› Predictive Analytics in Financial Industry
Learn how AI and predictive analytics are transforming finance, boosting revenue, and improving customer experience.
Oct 14, 09:30 AM UTC · 25 min · Presentation
› Real time ML at Roku
Learn how Roku scales real-time feature serving and speeds ML model development with Chronon
Oct 14, 09:55 AM UTC · 25 min · Presentation
› Bridging Real-Time and Batch: Declarative Feature Engineering with Apache Hamilton + Narwhals
Learn how Apache Hamilton and Narwhals unify real-time and batch feature generation for reliable, low-latency AI.
Oct 14, 10:30 AM UTC · 20 min · Presentation
› How Coinbase Builds Sequence Features for Machine Learning
Learn how Coinbase deploy user action sequences at scale to improve ML model performance.
Oct 14, 10:50 AM UTC · 20 min · Presentation
› Real-time ML: Accelerating Python for Inference (< 10ms) at Scale
Learn how a symbolic Python interpreter converts Python into DAGs to accelerate ML pipelines.
Oct 14, 11:10 AM UTC · 20 min · Presentation
› Real-Time Feature Aggregation at Scale: iFood’s Path to Sub-Second Latency
Learn how iFood built a sub-second feature platform with Spark and Redis to power real-time ML pipelines.
Oct 14, 11:30 AM UTC · 20 min · Presentation
› Building a Generative Recommender with Chronon
Learn how Chronon powers scalable, low-latency feature pipelines for next-gen recommendation systems.
Oct 14, 11:50 AM UTC · 15 min · Presentation
› On-Demand Feature Life Cycle Management
Discover Clicklease’s journey to building on-demand transformation functions for ML — powered by Hopsworks.
Oct 14, 12:05 PM UTC · 10 min · Wrap up
› Wrap Up
Jim Dowling wraps up the event.