Podcaster Untrained in Tech

Not a programmer? With A.I., just having a rough idea can be enough.

BACKGROUND

Journalist / Podcaster Kevin Roose (KR) introduces himself: “I am not a coder. I can’t write a single line of Python, JavaScript or C++. Except for a brief period in my teenage years when I build websites and tinkered with Flash animations, I’ve never been a software engineer, nor do I harbor ambitions of giving up journalism for a career in the tech industry. And yet, since late 2024, I’ve been coding up a storm!”

VIBE-CODING

‘Vibe-coding’ (a term popularized by the A.I. researcher Andrej Karpathy) is a useful shorthand term for the way that today’s A.I. tools allow even nontechnical hobbyists to build fully functioning apps and websites, just by typing prompts into a text box. You don’t have to know how to code to vibe-code – just having an idea, and a little patience, is usually enough.

“It’s not really coding,” Mr. Karpathy has said, “I just see stuff, say stuff, run stuff and copy-paste stuff, and it mostly works.”

Adds KR: “Among my creations: 

• A tool that transcribes and summarizes long podcasts
• A tool to organize my social media bookmarks into a searchable database
• A website that tells me whether a piece of furniture wil fit in my car’s trunk
• An app called LunchBoxBuddy, which analyzes the contents of my fridge and helps me decide what to pack for my son’s school lunch

These creations, useful only to a specific person who set them up, are all possible thanks to artificial intelligence (AI) and a new trend known as ‘vibe-coding.

KR’s own vibe-coding experiments have been aimed at making what he calls “software for one” – small, amateur-made, apps that solve specific problems in his life. These aren’t the kind of tools a big tech company would bother to build because there’s no large commercial market for them, their features are limited and some of them only ‘sort of work.’

But building software this way – describing a problem in a sentence of two, then watching a powerful A.I. model go to work building a custom tool to solve it – is a mind-blowing experience. It produces a feeling of A.I. vertigo, similar to what he felt after using ChatGPT for the first time. And it’s the best way he’s found to demonstrate to skeptics the abilities of today’s A.I. models, which can now automate big chunks of basic computer programming and may soon be capable of similar feats in other years. 

A.I. coding tools have existed for years. Earlier ones, like GitHub Copilot, were designed to help professional coders work faster, in part by finishing their lines of code the same way that Chat GPT completes a sentence. You still needed to know how to code to get the most out of them, and step in when the A.I. got stuck.

But over the past year or two, new tools have been built to take advantage of more powerful A.I. models that enable even neophytes to program like pros.

These tools, which include Curso, Replit, Bolt and Lovable, all work in similar ways. Given a user’s prompt, the tool comes up with a design, decides on the best software packages and programming languages to use, and gets to work building a product. Most of the products allow limited free use, with paid tiers that unlock better features and the ability to build more things.

To the non-programmer, vibe-coding can feel like sorcery. After you type in your prompt, mysterious lines of code fly past, and a few seconds later, if everything goes well, a working prototype emerges. Users can suggest tweaks and revisions, and when they’re happy with it, they can deploy their new product to the web or run it on their computers. The process can take just a few minutes, or as long as several hours, depending on the complexity of the project.

BUILDING AN APP

Asked to build an app that could pack a school lunch for a child, based on an uploaded photo of the contents of KR’s fridge: First, the app analyzed the task and broke it down into component parts. Then it got to work. It generated a basic web interface, chose an image recognition tool to identify the foods in the fridge and developed an algorithm to recommend meals based on those items.

If the A.I. needed KR to make a decision – whether he wanted the app to list the nutritional facts of the foods it was recommending, for example – it prompted KR with several options. Then it would go off and code some more. When it hit a snag, it tried to debut its own code or backed up to the step before it had gotten stuck and tried a different method.

Roughly 10 minutes after KR had entered his prompt, LunchBoxBuddy – which is what the A.I. had decided to call his app – was ready, it suggested a generic turkey sandwich. 

(Editor’s question: Assuming no daily changes in the content of KR’s fridge, would his personal A.I.-created app suggest another generic turkey sandwich every day?)

CHALLENGE – A.I. MISTAKES

Not all KR’s vibe-coding experiments have been successful. He was struggling for weeks to build an “inbox pilot” tool capable of responding to his emails automatically, in his writing style. 

He encounters roadblocks when trying to integrate A.I. work flows into apps like Google Photos and iOS Voice Memos, which aren’t designed to play well with third-party add-ons.

And, of course, A.I. occasionally makes mistakes. Once, when KR tried to build a website for a local tire shop, the A.I. made up fake reviews from the shop’s Yelp page and added them to a testimonials page. Another time, when KR tried to turn a long story he had written into an interactive website, the A.I. included about half the text and left out the other half.

Vibe-coding, in other words, still benefits from having humans overseeing the robots, or at least hovering nearby. And it’s probably best for hobby projects, not essential tasks.

That might not be true for much longer. Many A.I. companies are working on software engineering (non-human) agents that could fully replace human programmers. Already, A.I. is achieving world-class scores on competitive programming tests, and several big tech companies, including Google, have outsourced a large chunk of their engineering work to A.I. systems.

FUTURE OF UNTRAINED CODING

If KR were a junior programmer – the kind A.I. appears most likely to replace – he might be panicking about his job prospects. But KR claims to be “just a guy who likes to tinker, and to build tools that improve my life in small ways.” So, vibe-coding – or actual coding – is one area where A.I. is unmistakably improving itself and will thus improve his life.

Since talking about his vibe-coding experience on a recent podcast, KR has heard from dozens of other people who have been building their own personal tech tools with A.I. assistance. Colleagues have told him about the nutrition apps they’ve built to help them stick to their diets, or the apps they’re using to summarize the email newsletters they get. Readers have sent in websites they’ve built to track the price of eggs, or scrape Zillow listings in Los Angeles to discover instances of rent-gouging after the recent California wildfires.

CHALLENGE – POTENTIAL A.I. BAD RISKS

KR claims to be not naïve about A.I. or blind to the effects that A.I. coding apps could have on society if they continue to improve. He thinks it’s possible that an A.I. that automates building useful software could also automate the creation of a malicious code or even lead to autonomous cyberattacks. And he worries that software engineering is just the first office-based profession to experience the labor-replacing effects of A.I. tech tools.

VIBE-CODING GOOD POINTS

For now, KR believes that building apps to automate annoying or time-consuming tasks in his life seems as good a use of A.I. as any. So, he’s going to keep vibe-coding – at least until his son can pack his own lunch. 

__________________________________________________________

This career-related news story was based on an article written by Kevin Roose, published by The New York Times on March 4, 2025.