This post describes how we successfully migrated our serverless offering from AWS US-East to OVHCloud North America, reducing our monthly spend from $8,000 to $3,000 with no loss in service quality.
MLOps fallacies can slow AI deployment and add complexity. This blog breaks down 10 common misconceptions, their impact, and how to avoid them to build scalable, production-ready AI systems efficient.
Nearly a year ago, the Hopsworks team embarked on a journey to migrate its infrastructure to Kubernetes. In this article we describe three main pillars of our Kubernetes migration.
We describe the capabilities that need to be added to Lakehouse to make it an AI Lakehouse that can support building and operating AI-enabled batch and real-time applications as well LLM applications.
We present how Hopsworks leverages its time-travel capabilities for feature groups to support reproducible creation of training datasets using metadata.
This article introduces a taxonomy for data transformations in AI applications that is fundamental for any AI system that wants to reuse feature data in more than one model.
We present a unified software architecture for batch, real-time, and LLM AI systems that is based on a shared storage layer and a decomposition of machine learning pipelines.
In this post, we will look at how to put feature pipelines into production using Hopsworks.
In this article we describe the software factory approach to building and maintaining AI systems.
Hopsworks has added support for approximate nearest neighbor (ANN) indexing and vector similarity search for vector embeddings stored in its feature store.
At Hopsworks the F.A.I.R principles have been a cornerstone of our approach in designing a platform for managing machine learning data and infrastructure.
Learn how Hopsworks achieves high availability for the Online and Offline Feature Store and a Metadata layer, making it an operational ML system both for on-premise and cloud infrastructure.
This blog explores MLOps principles, with a focus on versioning, and provides a practical example using Hopsworks for both data and model versioning.
An ML model’s ability to learn and read data patterns largely depend on feature quality. With frameworks such as FeatureTools ML practitioners can automate the feature engineering process.
Discover the power of feature stores in modern machine learning systems and how they bridge the gap between model development and production.
In this article, we outline how we leveraged ArrowFlight with DuckDB to build a new service that massively improves the performance of Python clients reading from lakehouse data in the Feature Store
Explore the convergence of MLOps and DevOps. Learn about their purposes, differences, and areas of integration and discover best practices, challenges, and their future potential.
Find out how to use Flink to compute real-time features and make them available to online models within seconds using Hopsworks.
In this blog, we discuss the state-of-the-art in data management and machine learning pipelines (within the wider field of MLOps) and present the first open-source feature store, Hopsworks.
In this blog we present an end to end Git based workflow to test and deploy feature engineering, model training and inference pipelines.
Learn more about how Hopsworks stores both data and validation artifacts, enabling easy monitoring on the Feature Group UI page.
Discover how you can easily make the journey from ML models to putting prediction services in production by choosing best-of-breed technologies.
Learn how the Hopsworks feature store APIs work and what it takes to go from a Pandas DataFrame to features used by models for both training and inference.
Hopsworks Serverless is the first serverless feature store for ML, allowing you to manage features and models seamlessly without worrying about scaling, configuration or management of servers.
Hopsworks is the first feature store to extend its support from the traditional Big Data platforms to the Pandas-sized data realm, where Python reigns supreme. A new Python API is also provided.
Operational machine learning requires the offline and online testing of both features and models. In this article, we show you how to design, build, and run test for features.
We introduce how to use the What-If Tool as a Jupyter plugin on Hopsworks to build better machine learning models by making it easier to ask counterfactual questions about your model’s behaviour.
With support to Apache Hudi, the Hopsworks Feature Store offers lakehouse capabilities to improve automated feature pipelines and training pipelines (MLOps).
Read about how the Hopsworks Feature Store abstracts away the complexity of a dual database system, unifying feature access for online and batch applications.
Learn how to train a ML model in a distributed fashion without reformatting our code on Databricks with Maggy, open source tool available on Hopsworks.
This tutorial will show an overview of how to install and manage Python libraries in the platform.
Use open-source Maggy to write and debug PyTorch code on your local machine and run the code at scale without changing a single line in your program.
Evolve your models from stateless AI to Total Recall AI with the help of a Feature Store.
Connect the Hopsworks Feature Store to Amazon Redshift to transform your data into features to train models and make predictions.
In this blog, we describe how we leverage the authentication and authorization support in Open Distro for Elasticsearch to make elasticsearch a project-based multi-tenant service in Hopsworks.
Use JOINs for feature reuse to save on infrastructure and the number of feature pipelines needed to maintain models in production.
The feature store is a data warehouse of features for machine learning (ML). Architecturally, it differs from the traditional data warehouse in that it is a dual-database.
Hopsworks supports machine learning experiments to track and distribute ML for free and with a built-in TensorBoard.
Try out Maggy for hyperparameter optimization or ablation studies now on Hopsworks.ai to access a new way of writing machine learning applications.
Learn how to integrate Kubeflow with Hopsworks and take advantage of its Feature Store and scale-out deep learning capabilities.
We have many conversations with companies and organizations who are deciding between building their own feature store and buying one. We thought we would share our experience of building one.
This blog introduces the Hopsworks Feature Store for Databricks, and how it can accelerate and govern your model development and operations on Databricks.
Read how ExtremeEarth brings Large-scale AI to the Earth Observation Community with Hopsworks, the Data-intensive AI Platform.
This blog introduces platforms and methods for continuous integration (CI), delivery (CD), and training (CT) with ML platforms, with details on how to do CI/CD MLOps with a Feature Store.
Deep learning is now the state-of-the-art technique for identifying financial transactions suspected of money laundering. It delivers a lower number of false positives and with higher accuracy.
.svg)
.png)