Serverless Superhero

Our serverless superhero this week is Luca Mezzalira, principal serverless specialist SA at AWS. Many of us know Luca in one way or another, either through seeing him give a talk at a conference, reading his books, or through his blog posts. Luca champions micro frontends, among other things, and has become quite the voice in the serverless community. Thank you for everything you do, Luca!

Tutorials and Interesting Reads

I love that Lee Gilmore has turned his attention toward generative AI. His work is always well thought out and thorough, and the generative AI blog posts are no exception. Last week he taught us all about automating tasks using Bedrock Agents and AI. Lee walks through a meaningful example, tells us why it’s beneficial, and explains the architecture and workflow of agents. This is one of the best pieces of content I’ve seen on agents so far and it makes me want to go build one right now.

Observability is one of the major pillars in serverless. We emphasize it a lot - for good reason. That’s why I found Ivan Burmistrov’s article so interesting on using wide events instead of metrics. The post laments about modern observability’s focus on metrics, logs, and traces when instead maybe we should be emphasizing something known as a “wide event.” Ivan goes on to explain how they do observability at Meta and really gets your gears turning on whether or not we’ve been focusing on the wrong thing.

Speaking of observability, some more Rust content landed last week from James Eastham. His video focuses on OpenTelemetry for serverless Rust applications and covers everything from which crates to use, challenges with OTel and Lambda, and how to create trace contexts. This is your one-stop video for all things Rust and OpenTelemetry!

Marcia Villalba posted another video in her Step Functions series last week about service integration patterns. She covers the major differences between synchronous jobs and the wait for callback pattern, both of which definitely have their place in different workloads. This is a great video to help you understand the different ways you can trigger longer-running jobs in your workflows.

Spotlight

The spotlight this week goes out to one of my favorite builders, Jimmy Dahlqvist, who recently released his serverless handbook. Jimmy published his inspiring story into serverless last week, sharing his adventures, accomplishments, and thoughts on serverless services. It’s a great read and a fitting spotlight for an amazing community member!

Podcasts

I had Luc van Donkersgoed on the Ready, Set, Cloud podcast talking about taking your side projects to production. The episode focuses on how you learn differently when you build something meaningful and used by other people. I’m a firm believer in this practice and am thrilled I share the same opinion of last year’s AWS Go Build award winner 💙

On the Real-World Serverless podcast, Yan Cui had Richard Davison on the show to talk about the new Low Latency Runtime (LLRT) for AWS Lambda. Richard is the creator of the runtime and talks about his motivation behind it and how he was able to make it so darn fast.

Not quite a podcast, but Luc van Donkersgoed made another appearance last week, this time on the #believeinserverless livestream. He gave his Standing on the Shoulders of Giants presentation, highlighting PostNL’s serverless journey and their motivation behind going all-in on serverless. It’s a fantastic story with some great conversation at the end.

Tip of the Week

I thought Matthieu Napoli’s project was really cool last week. He build an app that shows you how Lambda scales execution environments when hit with a burst of invocations. This is a great visual for nailing down how it works behind the scenes.

How do serverless apps scale?https://t.co/xfNoSOG7R9 ✨

I built this website to *show* how AWS Lambda scales in real-time.
Play with it and watch it scale in response to your requests. It's all live. pic.twitter.com/GI5MHYnjgf

— Matthieu Napoli (@matthieunapoli) February 22, 2024

New Releases

EventBridge API destinations now support content-type header customizations. This means you can include things like application/cloudevents+json as your content type instead of being forced to use application/json only.

Step Functions now has support for open workflow metrics. This allows you to see how many active standard workflows you have at the account level. It also enables you to see your open workflow limit, which is a nice touch.

Lambda improved the response time when configuring stream and queue-based event sources. Now when updating your event-source mappings for Kafka, DocumentDB, or MQ, the change will take effect within 90 seconds, down from the SLA of 15 minutes.

Last Words

If you haven’t joined the conversation yet, I highly recommend checking out the Believe in Serverless community. Hundreds of serverless enthusiasts are there engaging in topics ranging from CI/CD to event-driven architectures. The community also hosts free weekly live streams from members. It’s a great place to learn and share your experiences!

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!

Allen