# Deepstream DLR Plugin

In this Lab, we are going to explore running NEO compiled models in NVIDIA DeepStream pipelines.

This GitHub contains necessary materials/artifacts to finish the following three steps:

1. Prepare and compile people detecting model
2. Build shared library for SageMaker NEO specific nvinfer deepstream plug-in, and use this plugin to run the DeepStream application locally
3. Deploy and manage deepstream application using AWS IoT Greengrass

Please follow the workshop page to get the step-by-step workshop instructions.