🦸 Serverless Superhero

Our serverless superhero this week is Nadia Reyhani, AI product engineer at Trusst AI and AWS Community Builder. Nadia is an amazing public speaker and an outstanding advocate (and role model) for women in tech. Thank you so much Nadia, for your leadership and for sharing your knowledge with the community!

💯 Spotlight

It’s really cool to see in-depth technical explanations from people who just “get it.” This is exactly what we got from Seth Orell last week with part 3 in his How to Crater Your Database series where he covers normalization. This blog post wonderfully describes some low-level details about why data normalization doesn’t do all that well when you reach internet scale. I’m a big fan of content presented this way, and I think Seth did a fantastic job. We could all learn a little something from this post.

🔥 My Favorite Content

There’s been so much talk on the model context protocol (MCP) in recent weeks. However, I’ve yet to see a good write-up or example of how to use it in a serverless environment like Lambda. Until now, that is. Last week, Davide De Sio wrote a short post introducing a starter template for using Lambda to create an MCP server. The blog post itself doesn’t go over the code, but the GitHub repository is a goldmine if you’re looking to get started.

And speaking of MCP, I wrote an article last week about using Kestra to observe tools on your MCP servers. I built a POC showing you how to capture and visualize tool usage without going through a ton of work to build a UI or assemble a complex backend. I had never used Kestra before, which is an open source workflow engine. It has an interesting DX unlike what I’ve used before with Temporal or AWS Step Functions. Anyway, I had a lot of fun building that project, take a look!

I felt a little out of touch last week when I read an article from Jonas Scholz titled Serverless is a scam. I’ve been under the impression that the tech community has generally come to accept serverless tech as a tool to build modern software. While Jonas’ post is a one-sided, “probably hasn’t used serverless in production” take on serverless - which we’re used to seeing these days - the amount of people agreeing with him in the comments was surprising. Serverless has its use cases and containers have their use cases, and most modern software can and should take advantage of both. But this article left me scratching my head on whether the community has reached that conclusion.

💡 Tip of the Week

With all the craziness going on right now in tech, it’s easy to become flustered or lose yourself with information and fear mongering. I was real happy to see Chris Williams’ post last week as an affirmation of some sorts.

🐣 New Releases

There was a crazy amount of announcements and releases from Cloudflare during their developer week. Rather than listing them all here, I will link to a great summary from Ashley Peacock that recaps everything that was announced.

S3 Tables got server-side encryption with customer-managed keys. This is the second week in a row with a big CMK release.

Lambda supports inbound IPv6 connectivity over PrivateLink now. It used to only support IPv4 or IPv6 over the public internet.

Last Words

It was kind of a light week for content last week, but that’s ok! Things are still moving, and I’m seeing a lot of content from first-time creators being produced 💙 If you ever need help creating, proofreading, or with ideas, please let me know - I’d be happy to help.

Happy Easter to those who celebrate 🐇 🐣

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