Microsoft Azure

MLOps in Azure ML with Iguazio

Utilize Microsoft Azure’s ML and a wide range of other data science solutions with an accelerated and automated way to rapidly deploy AI applications using the Iguazio MLOps Platform. Continuously deliver machine learning to production with the Iguazio-Microsoft joint solution for MLOps in Azure ML.


Using Azure and Iguazio together, data scientists, data engineers and DevOps teams can:


  • Deploy AI Models Faster: Build models on Azure ML and easily deploy them in production as scalable functions running on a scalable inference layer.
  • Monitor Models and Detect Drift: Leverage Iguazio’s monitoring functionality to identify drift in models developed on Azure. When drift is detected, Iguazio automatically triggers a training process in Azure ML to optimize the model.
  • Utilize a Feature Store: Use Iguazio’s feature store in Azure to reduce the development cycle and the overhead of feature engineering. Data scientists can write online and offline features once and use them for both training and online inference.
  • Optimize Feature Management: Create and manage features in the Azure ML environment with Iguazio’s feature store. These features can be used for both online inference and offline training.
  • Support Real-time Use Cases: Create real-time ML pipelines by leveraging Azure Event Hubs along with Iguazio’s functions and robust feature engineering capabilities to support real-time ML use cases such as predictive maintenance, real-time recommendation engines and fraud prediction.

Seamlessly integrates with:

  • Azure Machine Learning
  • Azure DevOps
  • Azure Kubernetes Service
  • Azure Blob Storage
  • Azure Synapse Analytics
  • Event Hubs
  • Azure Stack

Accelerate and Automate MLOps with Iguazio & Azure ML

Continuously deliver machine learning to production with the Iguazio MLOps Platform and Feature store in Azure ML.

Explore More of the Azure – Iguazio Partnership

For inquiries on the joint Microsoft Azure-Iguazio solution please contact us: