In this we are going to start with the twitter api and use node.js in aws sdk to create program to hold tweets, from our node js applications we are going to feed the tweets directly to the Kinesis stream, which then pipes them into the kinesis firehose, then firehose is configured to drop them in s3 bucket, then we are going to use a lambda routine to unpack all the raw tweets and land tem in the dynamodb table and then we are going to use a dynamodb stream from that table to trigger another lambda routine which will call the amazon comprehend API returning the sentiment for the tweet, we are going to use that result to increment counter it on another dynamodb table to tally the sentiment of the text.