Hi @KristheTech, to make this work, you'd need to do a few things: You'd need to read the image data and store it into a buffer the size of the resolution and image density. Send the image data as a MQTT message (you can use the Blinky Hello World example). Keep message at/under 128kb. Set up an AWS IoT rule to save each message into an S3 bucket. Have Amazon Rekognition read from that same S3 bucket. Do something with the output from Rekognition.