Here’s a link to the episode…

It’s scary good.

He came on my podcast to talk about how he’s actually using AI in marketing, and:

• Why he moved from ‘big company CMO’ to a scrappy team of 5. 
• How they’re using AI to drive pipeline right now.
• How to build a culture of AI adoption. 
• His formula: Marketing Output = AI × Taste²
• Where the AI puck is going.
• And of course, how he’s replacing himself with his Claude Code AI CMO.

Here’s a link to the episode.

Let me be clear: no one asked for this. But I’m going to walk you through how I built and published an iOS app using Claude Code, Xcode, and pure vibes.

I don’t know Swift. I’ve never written a real line of Swift in my life. But I’ve now shipped four apps to the App Store, mostly by telling AI what I wanted and watching it make the magic happen.

This post isn’t a tutorial in the traditional sense. It’s a field report from the future, where anyone with an idea and some time can build software without writing code.

Let’s get into it.

🧰 What You Need (It’s Not Much)

You’ll need just two things to get started:

• Claude Code – Download it from Anthropic’s site.
• Xcode – Apple’s all-in-one IDE for building and submitting apps.

That’s it. No Swift knowledge. No CS degree. No excuses.

🧪 My App Idea: A Data Sheet, But Make It an App

I hate product data sheets. PDFs that no one reads, packed with boilerplate and buzzwords. What if, instead, you could open a beautiful, interactive app that explains what your product actually does?

That’s what I set out to build: an interactive data sheet for incident.io, the company where I’m CMO.

So I created a new iOS project in Xcode and named it (very creatively) incident-io-datasheet. Then I opened Claude Code, pointed it at my project folder, and told it what I wanted:

“Build a beautiful, interactive iOS app that showcases the four products incident.io offers. Make it look like Apple made it. It should be WWDC-award-worthy.”

A tall order. But Claude said “Yes.”

🔧 Vibe Coding in Real Time

Here’s what happened next:

1. Claude scanned the project and figured out I was building an iOS app in Swift.
2. It asked to initialize the Claude project and created a claude.md instruction file.
3. I gave it the prompt above. It broke the job into 5 steps: research incident.io, design the app, create the data model, build a Swift UI interface, and add interactivity.

It wrote real Swift code. Created product cards. Designed a sidebar. Built the views. Even added animated transitions.

I watched the whole thing unfold like it was Christmas morning. No clue what the code did, but it looked legit.

🐛 Fixing Errors (The Lazy Way)

Like any good dev, Claude got some stuff wrong. Build failed. A bunch of red errors in Xcode.

But unlike normal devs, I didn’t debug anything. I just copied the error messages into Claude. It fixed them.

Rinse. Repeat. Eventually: build succeeded.

Then came the moment of truth: I hit play in the iOS simulator.

And there it was: a real app.

📱 The App Actually Worked (?!)

It had:

• A home screen with our tagline: “Move fast when you break things.”
• Tabs for each product (Alerting, Response, Status Pages, Postmortems)
• Beautiful floating bullet points with key features and benefits
• A sidebar nav that just… showed up on iPad, like it knew it belonged there

I even told Claude to add a flame emoji next to the company name. It tried. It was the wrong emoji, but hey, it tried.

📤 Publishing to the App Store

Building the app was surprisingly easy.

Getting it on the App Store? Slightly trickier, but still totally doable.

Here’s what I had to do:

• Join Apple’s Developer Program ($99/year)
• Create an App Store Connect record (name, description, screenshots, pricing)
• Archive the app in Xcode
• Generate and add an icon (Claude vibe-created a 1024x1024 flame emoji PNG—somehow perfectly sized on the first try)
• Fix minor metadata and compliance things
• Submit for review

It took about 45 minutes total to go from working app to “Waiting for Review.”

🏁 Final Thoughts: You Can Actually Do This

Vibe coding is real.

I didn’t know Swift. I barely touched Xcode. I just told Claude what I wanted and helped it fix the errors it created along the way. The end result? A working, beautiful app that’s now on the App Store.

You can do this. Whether it’s a data sheet, a dumb game, or a side hustle you’ve been sitting on… Claude Code and Xcode are enough to get something shipped.

And if you’re lucky, maybe your flame emoji app will win a WWDC award. Or at least confuse a few Apple reviewers.

It’s long, so get comfortable.

####

In this week’s live event, Chris is joined by Tom Wentworth, CMO of Recorded Future, to discuss the intricate collaboration between marketing leadership and finance. They dive deep into the often-overlooked aspect of bringing business acumen into financial planning, emphasizing the importance of marrying marketing metrics with business objectives to drive enterprise value.

*Tom emphasizes the importance of narrowing down account focus, even for large enterprises, suggesting that a smaller, targeted set of accounts can yield better results. They dig into practical challenges and solutions in demand creation, attributing their lower cost per opportunity to a more strategic allocation of marketing spend. Tom’s advocacy for direct mail as a unique, effective tactic highlights an often-overlooked channel. Moreover, the value of robust financial relationships, notably with the CFO, and the innovative use of a news organization to drive engagement and brand authority are discussed in depth.* \

This episode features an interview with Tom Wentworth, CMO at Recorded Future, the world’s largest threat intelligence company, working with over 1,600 businesses and government organizations across more than 74 countries.

In this episode, Tom talks about building Recorded Future News, a separate P&L that has become one of the largest cyber intelligence news sites, and a trusted media entity that breaks important news in the cybersecurity/intelligence world.

Key Takeaways:

• Brand building is not just about creating awareness, it is about building trust and building relationships. 
• Sometimes, the only information you need to determine ROI is whether your clients and prospects feel your initiative provides them with value.
• A quick way to fail as a CMO is to run someone else’s playbook. Make sure that you understand the current company and context; create a bespoke plan. 

Quote:  “If you think about brand building, what most people do in cyber security, when they get to be Recorded Future size, we’re about over 300 million in ARR, which makes us a pretty good sized company. What most companies our size do, is they go buy an F1 sponsorship, or they go put their logo on some sports team’s jersey. And that’s cool. That’s a cool way to get in front of people, but I don’t think it builds trust. Like, why would I trust this vendor just because their logo is on a jersey or their logo is on a car? That just tells me they’ve got a lot of money to spend. So we said, instead of going to do that, the way that we’re going to build awareness for Recorded Future and build a relationship with our audience is through news and then through a podcast, which I’ll talk about a little bit too. So that’s our brand awareness investment. We feel very confident that people who fall in love with our news are also going to like our product. I don’t have to prove it, in the sense that I don’t have to have this super complicated attribution scheme. We just ask our clients and our prospects, ‘Hey, do you like Recorded Future News?’ And they love it. That’s just about enough for us.”