During re:Invent, AWS announced a new feature within AWS Lambda called durable functions. These are the same Lambda functions we all love, but they let you run multi-step workflows by keeping checkpoints and state. What does this mean? You can run similar functionality to what we’ve typically used AWS Step Functions for. But instead of using Amazon State Language, you can use familiar code and dependencies.

The Lambda Powertools team released a new feature that allows your Lambda functions to buffer logs. That sounded cool, but I didn’t understand how it would work or how logs would show in CloudWatch. I decided to try it out and provide a visual example of how it looks with different configuration options.

On my post called Secure API Gateway with Amazon Cognito using SAM I talked about different Auth terms and walked through a setup to use the Client Credentials Flow, but Cognito recently introduced pricing changes for machine-to-machine authentication that will make this cost us and my main goal is to do this while staying in the free tier for personal projects that will not be generating any income. That is why in this post I am going to setup Amazon Cognito using a different flow called user…

When you have CI/CD pipelines that run automated tests against your APIs you might need to dynamically create users in Amazon Cognito to run them. If that is the case you are in the right place. In this post we’ll be going over what you need to do to create a valid user in Cognito to be used by your automation.

I create a lot of APIs, these are for blog posts, for playing around with new functionality or tools that I’ve created for myself. All of these have been created without authentication in place. Not securing APIs can create data exposures for you, but it can also pose a financial risk to your accounts if a malicious user gets your endpoints. This is why I want to secure any API I create but I want this to be with minimal setup so that it’s simple to replicate many times.