Backstage interview
summary
A European full‑stack AI platform called Hopsworks claims to provide a feature store and real‑time LLM serving that runs on Kubernetes, is deployable with helm install, and manages GPU clusters on‑premise. The demoable specifics are GPU scheduling details—priority, preemption, utilization optimization—and an assertion it supports batch, realtime and LLM workloads at scale. A book and a global Feature Store Summit are announced, suggesting a product roadmap and community traction.
AI-written summary · about
related talks
01h 10mReal-time AI Lakehouse
33mKeynote: Federated Analytics at Scale
31mFeast as a Cloud Native Data System: Bridging Batch Processing and Online Serving...
38mQuerying Live Data With LLM App
23mBuilding Production-Ready Data Pipelines for AI/ML with Lambda Architecture
40mBuilding highload ML powered service
30mPython and the AI Lakehouse
27mStream, Store, Visualize: Fast-Track Real-Time Analytics the OSS Way