AI Connect | Data Science Connect
AI ConnectExplore the 2027 portfolio

ML Operations

Hopsworks

Feature store and AI lakehouse platform for ML pipelines and real-time serving

Visit website ↗See if it’s a matchRequest an introduction

About Hopsworks

Hopsworks provides a feature store with low-latency online retrieval, plus an AI lakehouse and MLOps tooling for experiment tracking, a model registry and deployment. It supports cloud, on-premises, air-gapped and hybrid setups, and is developed by Hopsworks AB in Sweden.

Common use cases

  • Real-time fraud detection
  • Recommendation features
  • Feature reuse across teams
  • Sovereign AI deployments

Products and services

Feature Store

Central repository for feature data with sub-millisecond online retrieval powered by RonDB.

AI Lakehouse

Works with Delta, Iceberg and Hudi tables via a Python-native query engine.

MLOps Platform

Experiment tracking, model registry and deployment pipelines.

Sovereign AI

Blueprint for air-gapped, on-premises or hybrid deployments.

Key features

  • Online and offline features
  • Point-in-time correctness
  • Model registry
  • Ray training
  • KServe and vLLM serving
  • Air-gapped deployment

Integrations

  • Delta Lake
  • Apache Iceberg
  • Hudi
  • Ray
  • KServe
  • vLLM
  • Snowflake
  • Databricks

Hopsworks alternatives

FAQs

Is Hopsworks open source?

The core platform has open source roots; commercial editions add enterprise features.

Where can it run?

Cloud sign-up, on-premises, air-gapped or hybrid.

How is latency achieved?

Online feature retrieval is powered by RonDB, per the vendor.

Work at Hopsworks? Request updates to this profile. Update this profile

Compare with similar providers

All ML Operations providers