logicalclocks/hopsworks-tutorials

What it solves

This tutorial collection helps developers and data scientists learn how to build machine learning systems using Hopsworks, a feature store platform. It covers everything from basic quickstarts to advanced use cases like real-time fraud detection, batch loan approvals, and LLM-based document Q&A, while also demonstrating integrations with popular tools.

How it works

The tutorials are Jupyter notebooks that you run inside Hopsworks after creating an account. They show you how to log in, connect to the feature store, and perform tasks like feature engineering, model training, and deployment. Each tutorial is organized by category: real-time AI systems, batch AI systems, LLM AI systems, API examples, and integrations with third-party services.

Who it’s for

Data scientists and ML engineers who are already familiar with Python and machine learning concepts and want to apply them in a production environment with a feature store. Also useful for anyone exploring Hopsworks and its capabilities.

Highlights

• QuickStart notebook to get started in minutes
• Real-time AI systems: fraud detection, anti-money laundering, TikTok-style recommendations, and timeseries prediction
• Batch AI systems: loan approval, fraud batch, churn, credit scores, hospital wait time, and taxi fare prediction
• LLM AI systems: fraud cheque detection, RAG-based PDF assistant, and fashion recommender
• API examples for vector similarity search, datasets, feature groups, monitoring, Git, jobs, Kafka, OpenSearch, and more
• Integrations with Airflow, BigQuery, Snowflake, Spark, LangChain, and many others