Build AI Agent with function calling - EASY in AWS Bedrock

In this video, I’ll show you a simple TypeScript example of how to do function calling using AWS Bedrock. Our scenario will be pretty straightforward: we’ll have a function that fetches some updated in real time external data — like the currency exchange for example. Then we’ll orchestrate a call to that function through Bedrock. We’ll see how the model can respond with the correct JSON payload to call the function, and how we handle that in our TypeScript code. Ready? Let’s do it. 00:00 - Intro 01:00 - What is function calling? 01:52 - Plan for this video 02:50 - Function calling with Agent & AWS Lambda in AWS Console 03:50 - Create an AI Agent 06:50 - Add AWS Lambda function 13:00 - Test the AI Agent in AWS Console 15:10 - Call the AI Agent from your code 21:15 - Function calling with Converse API 46:59 - Demo Guthub repo: https://github.com/nim-f/bedrock-tutorial , branches 07-function-calling and 08-function-calling-agent Exchange rate API: https://www.vatcomply.com/documentation Documentation: https://docs.aws.amazon.com/bedrock/latest/userguide/tool-use-inference-call.html https://community.aws/content/2hW5367isgQOkkXLYjp4JB3Pe16/intro-to-tool-use-with-the-amazon-bedrock-converse-api