NVIDIA DeepStream development with Microsoft Azure

AI Engineer
Data Scientist
Azure IoT
Azure Container Registry
Azure IoT Edge
Azure IoT Hub

With NVIDIA DeepStream, you can seamlessly develop optimized Intelligent Video Applications that can consume multiple video, image, and audio sources. You can also apply single or cascading inference operations to video frames in real-time, and transmit inference results to the cloud for archiving or additional processing.

Learn how to develop custom DeepStream applications backed by services in Microsoft Azure and securely deploy NVIDIA DeepStream-based AI workloads to NVIDIA-powered IoT Edge devices.


  • Knowledge of containers
  • Familiarity with Azure IoT Edge
  • X86 Based System running Ubuntu 18.04 OS
  • RTX2080, RTX3080, T4, or A100 NVIDIA Graphics Card
  • NVIDIA Jetson Embedded Device provisioned with JetPack 4.6

Modules in this learning path

In this module, you'll learn how to set up and configure an x86-based Ubuntu 18.04 system to host an NVIDIA DeepStream development environment.

In this module, you'll learn how to set up and configure the NVIDIA DeepStream 6.0 Graph Composer on an X86-based Ubuntu 18.04 system to enable rapid development of Intelligent Video Analytics application pipelines for deployment to cloud and edge-capable devices.

In this module, you'll publish and deploy an ARM-based DeepStream container workload to NVIDIA embedded hardware by using Azure IoT Edge.