Mike Chambers from AWS teaches how to build and deploy serverless agentic applications using Amazon Bedrock. The course covers initializing agents, adding tools, code-execution guardrails, RAG, and deployment, using a customer-service chatbot for a mug business as the hands-on example.
I'm excited to introduce serverless agentic workflows with Amazon Bedrock, built in partnership with AWS and taught by Mike Chambers, who is a Senior Developer Advocate at AWS specializing in Genai. Mike also taught a previous course on serverless OLM apps with Amazon Bedrock.
It's my pleasure to be back here with you Andrew and teach this course.
Agentic workflows carry out workloads that can be hard to predict in advance depending on the user input. Maybe an agent will make a lot of some API calls, or maybe not. A serverless architecture lets you build something to handle these chaotic workflows in an efficient way. Importantly, this lets you get your agentic workflow deployed much faster. In this course you learn to build and deploy a scalable serverless agentic application into production. You initialize an agent, add tools, code execution guardrails and rag, and you deploy all the components of the application in a serverless environment. Using a service architecture means you don't have to worry about maintaining your own service. Your agent can make an API call say to do a web search and summarize the results or whatever it wants. And even if you didn't have a hot server that you're paying for ready to serve that query, the serverless architecture can take care of it. And so this is particularly important for agents of access to say a lot of tools and does a lot of potential APIs they might want to call at any moment. These concepts will be illustrated with examples from a customer service chat agent and RAG systems that might, for example, fetch some information and then decide it needs to go back and get more information. And a genetic workflow lets these algorithms decide when to fetch more information. And the serverless architecture lets you deploy such complex workflows quickly.
Right Andrew? In this course you will build a customer service bot for a fictional mug business through hands on exercises. You'll build the bot, connect multiple types of agent actions, and implement guardrails. You will do all of this using the fully managed Amazon Bedrock service. The course will emphasize two elements critical to the deployment of business applications. The first is serverless deployment, allowing rapid scaling and getting away from well it worked on my laptop. All the code you write in this course is configuring your agent in the cloud where it can run and scale. The second is responsible agents. You'll learn to protect your application from malicious prompts and to protect your business from unintended outputs by configuring guardrails.
So if you're considering building an agentic application for a business or for personal use. This course will help you go from your idea to running in deployment. Hope to see you in the course.
