Hey friends 👋 It’s been a minute, huh?

You might have noticed that I haven’t sent a new issue of the newsletter in quite a while. Six weeks, to be exact. Six!

So, first of all, sorry about that. I missed you all, and I hope some of you missed getting an email from me every other week.

I feel like I have some explaining to do. But before doing that, we need to talk about the elephant in the room. No, not that elephant. That’s one of my daughter’s giant plushies that’s been hanging out in my office because she ran out of space in her bedroom.

I’m talking about the other elephant in the room: AI.

It's mid-2025, and AI is still a polarizing topic. Some of you love it. Some of you would rather eat sand than hear more about it.

And I get it, which is why I’ve always tried to keep this newsletter as AI-neutral as possible. Sharing an “AI is great” article here, an “AI is bad” blog post there, and trying to keep the newsletter on the subject I like to think we are all interested in—how to use software design and architecture principles to build frontend applications at scale.

But over the past few months, keeping that neutrality became practically impossible. Not only because I was deep in the AI rabbit hole and was becoming increasingly enthusiastic about it, but because I felt that by intentionally not voicing my opinion on AI, I was keeping us out of a very important conversation happening all around us.

Whether we like it or not, AI is changing the way we build software these days, and at the very least we should figure out exactly how.

Now, what excites me about AI is not necessarily talking about how to vibe code a SaaS in an afternoon, or playing with the five hundred coding agents that come out every week. What I’m really interested in is exploring how AI can help us build better software that lasts longer.

The reason I ended up in the AI rabbit hole in the first place is that a question popped into my head and I wanted to figure out the answer: “How can AI help us build software both faster and better?”

So, after a few weeks of banging my head against various walls, I did what I apparently always do when I have a question I can’t stop thinking about: I started a newsletter. A Substack, specifically. It’s called Code Like You Mean It.

Code Like You Mean It is a place to hang out with other nerds who want to build software that lasts. It’s not an AI newsletter, but we will talk plenty about it as we explore exactly how it’s changing this practice of building software we all know and love.

Above everything else, Code Like You Mean It is a place for people who care about the code they write—whether they write it by hand or using an LLM—and who are willing to code with intention (hence the name).

To learn more about what CLYMI is all about (and join the hottest new club in town), check out the first issue. If you enjoy my software design ramblings over here, I’m sure you’ll find it interesting.

And if you do find it interesting, let me know in the comments! The community features of Substack, like comments and chats, are a big reason why I chose it for this new newsletter. I really hope the issues there can feel more like a conversation with you all rather than a one-sided monologue.

Alright, I see you have a question for me.

“The new Substack is great and all… but what about this newsletter, Maxi? Have you given up on frontend development?? Are you abandoning us here?? Are we breaking up?!”

Of course not! Frontend development is my happy place, and this newsletter is still one of my favorite things I’ve ever done, so no, we’re not breaking up!

Plus, while the new Substack is more “full-stacky”, I’m still a frontend developer at heart, and a lot of the content I read and the tech I play with is very much frontend and UI-related. Those, plus some of my usual ramblings, will continue to be part of Frontend at Scale.

The plan for the foreseeable future is to keep sending you my favorite frontend-related resources over here, at the usual bi-weekly cadence, and ramp up the new newsletter with an article every week or two.

And you know what they say about plans—they always go perfectly and nothing ever goes wrong.

Anyway, I don’t want to make this too long, so I’ll leave you with my favorite reads and watches from the last couple of weeks, and I hope to continue seeing you around here and at the new place.

It’s great to see you again, friends. Talk to you soon!

ARCHITECTURE SNACKS

Links Worth Checking Out

1. I really enjoyed this short, 12-minute lightning talk from Jake Albaugh on aligning code to design intent. It's a quick introduction to the Figma Dev Mode MCP, but more importantly, it's a great reminder of why design-engineering collaboration is so important. "Meaning makes things worth doing, and design is how you bring meaning into a product."
2. Gergely Orosz covered the fascinating story of an over-employed engineer who managed to get hired at at least a dozen different AI startups over the last few years (many of them YC-backed) and how he eventually got caught holding multiple jobs at the same time—sometimes even multiple in-person, full-time jobs!
3. Elena Verna wrote about the rise of the AI-native employee based on her experience at Lovable—a ~35-person startup that grew to $80M ARR in just 7 months. “Move fast and break things is back - but this time with 10x output and 1/10 the mess.”
4. And for some thoughts on the other side of the AI-spectrum, here's an article by Joshua Barretto on the joy of writing toy software by hand, without LLMs, just to learn and have fun and remember what our jobs used to feel like in the before times of 3 years ago.
5. Do you usually struggle with flaky and superfluous tests, or tests so slow that they prevent you from running the entire suite on every change? Well, according to André Arko, we should just delete those tests. You don't need to tell me that twice, André.
6. Jordan Cutler wrote about the five levels of communicating impact as an engineer, just in time for the start of everyone's favorite season—☀️performance review season.
7. Thorsten Ball was a guest on the Changelog podcast to talk about Amp (Sourgraph's coding agent), his widely popular How to Build an Agent article, how he uses AI to think better, and much more.
8. Prompt Engineering is so first half of 2025, the hottest new trend is Context Engineering. According to Philipp Schmid, "The secret to building truly effective AI agents has less to do with the complexity of the code you write, and everything to do with the quality of the context you provide."
9. ECMAScript 2025 was recently announced, including new helpers for Sets and Iterators, a new Promise.try() method, a bunch of RegEx features, and much more. Paweł Grzybek wrote a nice summary of everything that's new in the latest vesion of the spec.
10. ​Jake Archibald hates footnotes, and he wants us to hate them too.

That’s all for today, friends! Thank you for making it all the way to the end. If you enjoyed the newsletter, it would mean the world to me if you’d share it with your friends and coworkers. (And if you didn't enjoy it, why not share it with an enemy?)

Did someone forward this to you? First of all, tell them how awesome they are, and then consider subscribing to the newsletter to get the next issue right in your inbox.

I read and reply to all of your comments. Feel free to reach out on Twitter, LinkedIn, or reply to this email directly with any feedback or questions.

Have a great week 👋

– Maxi