Serverless Picks of the Week
Issue #42: Start Something New

Serverless Superhero

Our serverless superhero this week is Taavi Rehemägi, co-founder and CEO of Dashbird. Taavi is breaking the mold in serverless observability. He’s built an expertise on what does and doesn’t matter in monitoring a serverless application and has applied his knowledge in Dashbird with intelligent warnings and meaningful alarms. Thank you Taavi for helping us build stronger applications by giving us the tools to know what is really going on.


I keep getting the feeling that Rust is going to be big this year. If you’ve been considering using Rust for your Lambda functions, Cliff Crosland published an article on how to get started. He covers some of the benefits of using Rust (there’s more than just speed!) and gives a great breakdown of all the components needed to build a function. I hope we see more content like this in 2023.

We all know one of the hardest problems in software engineering is cache invalidation. When data changes, how do you guarantee your cache has the updated values within your SLA? Ellery Addington-White wrote an incredibly clever way to improve cache hit rates by using DynamoDB Streams. He built a middleware component that plugs into the AWS v3 JavaScript SDK using Momento for easy read-aside caching and another one that automatically updates a cache directly off a DynamoDB stream. This is the kind of innovation I love to see, and it’s only January!

A common pattern we see with AWS alarms is to use an SNS topic that triggers a Lambda function that posts to Slack. Davide de Paolis has come up with a better way to get your alarms in front of you. He proposes to send Slack notifications with AWS Chatbot instead of a Lambda function. He shows the difference in resources in a CDK script and explains the benefits of using ChatBot vs a function. This is a great way to optimize alarm infrastructure and opens the door for two-way communication in Slack with your AWS environment!

Interesting Reads

Twitter’s newsletter service, Revue, closes its doors this week, leaving many folks out of a solution - including me with this newsletter. I looked at several options to continue sending you high quality issues without service interruptions, but most alternatives are a bit pricey. So I (Allen Helton) decided to build my own serverless newsletter publishing app. In the post, I cover some of the features, why I chose the providers I did, and linked to the source code on GitHub. If you’re looking to start a newsletter, this is an easy way to start!

When it comes to DynamoDB data modeling, it’s not about the number of tables. Ex-AWS DynamoDB expert and Principal Solutions Architect at Momento Pete Naylor believes this so strongly, he actually advocates for retiring the term “single table design”. Instead, focus on using DynamoDB’s schema flexibility and item collections to optimize your model. The requisite number of tables will follow. Read his full thoughts on the Momento blog—and stay tuned for more! Sponsored

There was a lot of backlash when OpenSearch serverless was released at re:Invent. Many people jumped on it saying it isn’t actually serverless. That might be the case, but it still offers significant cost savings to consumers who are already using OpenSearch and have large instances and cluster sizes. Martyn Kilbryde goes through how to implement it in your serverless apps to start saving money today. There are some tricky things to look out for that Martyn points out and he offers a few good insights into the use case for the service.

New Releases

You can now set the maximum concurrency for SQS to Lambda integrations. This is a clean implementation to prevent runaway processing instead of setting reserved concurrency on your functions.

The Edge got a little boost last week as Lambda@Edge functions now support nodejs18.x!


One of the cloud development principles I preach regularly is serverless-first, not serverless only. This is an easy mantra to say but in practice can be deceptively difficult. Saying you are serverless-first is great for buzzword bingo, but there are changes that need to be made in the way you think, architect, build, and operate. Sheen Brisals wrote about the FIRST principles for a successful serverless adoption last week in a sobering article that reminds us that serverless development is a lot more than “hello world”. He covers the -first principles your org must go through before you can safely transform into a true serverless-first mindset.


The Ready, Set, Cloud podcast officially kicked off last week! In the first episode, I talk with Khawaja Shams about the true definition of serverless. We cover the impact of serverless on society and what we stand to lose of the definition continues to fall apart. It’s a great listen if you’ve been following the debates on social media.

Tip of the Week

I know not everyone is a fan of SAM (you should be, but I won’t rant). But a big update came out this week that adds tremendously to quality of life. You can now lint your templates via a terminal using the sam validate command!

Last Words

2023 is turning out to be a big year already. Lots of great features are coming online and new content is being created by so many of you wonderful builders. Thank you for choosing to build in public and share your findings with the community!

If you’d like to make a recommendation for the serverless superhero or for an article you found especially useful, send me a message on Twitter, LinkedIn, or email.

Happy Coding!


This Week's Sponsor

This issue is brought to you by Momento, a completely serverless cache system. Follow Momento on Twitter and LinkedIn to stay up to date.

Join the Serverless Picks of the Week Newsletter

Stay up to date with the best content serverless has to offer, learn about the latest updates to AWS serverless services, and get to know community superheroes, catered by AWS Serverless Hero Allen Helton. New issue every Monday.
Click here to see past issues.

Join the Serverless Picks of the Week Newsletter

Thank you for subscribing!
View past issues.