What happened when I decided a laptop wasn't a requirement anymore
It started on a Tuesday night, the kind where I was lying in bed with my phone propped against a pillow, scrolling through other people's beautifully designed personal sites and feeling that familiar itch. I didn't have a laptop with me. I had a phone, a data plan, and a growing impatience with the idea that "real" building requires a desk, a mouse, and hours of uninterrupted focus.
So I asked myself a question I hadn't taken seriously before: could I actually build a website, from scratch, using nothing but my phone and an AI tool to guide me through it?
I decided to find out.
The decision that felt small until it wasn't
I've spent years writing for other people, shaping their ideas into words that sell, persuade, and build trust. But my own website had been sitting in my head for months, half-planned, fully imagined, never built. Every time I thought about starting, I pictured myself hunched over a laptop for a weekend, wrestling with code I didn't fully understand.
That image kept the project stuck in "someday."
What changed wasn't a burst of motivation. It was smaller than that. I simply opened a chat window on my phone and typed out what I wanted: a warm, editorial site with a navy and gold palette, serif headings, clean body text, something that felt like it belonged to a writer who takes her work seriously but doesn't take herself too seriously.
I wasn't sure it would work. I half expected to hit a wall within the first twenty minutes.
The first real surprise
The first thing that struck me wasn't the code. It was how much the process resembled a conversation with a collaborator who happened to know things I didn't. I described what I wanted in plain language, the way I'd explain a vision to a designer over coffee. The AI translated that into structure: components, layout logic, color tokens, typography choices.
I didn't write a single line of code myself. I typed sentences. Full ones, with feeling in them. "I want the header to feel like the opening line of a good essay, not a billboard." And somehow, that instruction turned into something real on the screen.
Typing on a phone for something this technical should have been miserable. It mostly wasn't. Voice-to-text carried a lot of the weight, and when I did type by hand, the back-and-forth was short enough that my thumbs didn't rebel.
Where it got hard
I won't pretend it was seamless the whole way through. There were moments where what appeared on screen didn't match what I had pictured. A section that should have felt spacious came out cramped. A shade of gold that looked warm in my head rendered too loud on the small screen.
Each time, I had to do the thing I do professionally for other people but rarely do for myself: name exactly what wasn't working and why. Not "this looks wrong," but "this section feels too tight, give it more breathing room, and pull the gold back so it reads as an accent instead of a shout."
That precision mattered more than I expected. Vague feedback got vague results. Specific feedback, the kind I'd normally reserve for a client brief, got me exactly what I was picturing.
There was also the plain physical strain of building something visual on a six-inch screen. Previewing layouts, checking spacing, judging whether a heading felt balanced against the paragraph below it, all of that asked more of my eyes and patience than I expected going in.
The part that surprised me most
Somewhere around the third or fourth round of edits, I stopped thinking of this as a technical experiment and started thinking of it as an act of authorship. I wasn't managing software. I was directing something, the way I'd direct a piece of long-form writing, paragraph by paragraph, until the shape of it matched the shape in my head.
That reframing changed how I approached the rest of the build. I stopped asking "can AI build a website" and started asking "what do I actually want this space to say about me and my work." The tool was never really the hard part. Knowing what I wanted, clearly enough to describe it, was the actual work.
I think that's true of most creative collaborations, human or otherwise. The tool amplifies clarity. It doesn't manufacture it.
What it felt like when it finally looked right
There was a specific moment, late at night again, when the homepage finally matched what I'd been carrying in my head for months. Navy background settling into itself, gold catching the eye without overwhelming it, the serif headings giving the whole thing a quiet, literary confidence.
I sat with that for a while before doing anything else. Not editing, not tweaking, just looking at it. It felt less like I'd finished a task and more like something I'd been describing to people for months had finally stopped being an idea and started being a place.
I built that from my bed, in fragments, between other parts of my day, using a phone I also use to answer emails and scroll through news I probably shouldn't read before sleeping.
What I'd tell someone considering this
If you're waiting for the "right setup" before you start building something you care about, I'd gently push back on that instinct. The barrier was never really the laptop. It was the belief that building requires a certain kind of environment, a certain kind of expertise, a certain kind of permission.
What actually got the job done was clarity, patience, and a willingness to describe what I wanted in painfully specific terms instead of vague hopes. The phone was just the surface I happened to be holding when I stopped waiting.
I still don't think a phone is the ideal tool for this kind of work. There are moments where a bigger screen and a proper keyboard would have saved me real time. But it proved something I needed proving: the excuse of not having the "right" setup was mine to let go of, and I finally did.
The site exists now. Navy and gold, serif and quiet confidence, built in pieces from a bed I wasn't planning on leaving that night. Some things start smaller than you expect, and turn out bigger than you thought they could.
Top comments (8)
I have been building websites using my smartphone 10 years ago. Using CMS, to be fair. But I also had a UNIX console and Python installed back then.
As others wrote already, we'd be excited to see your website screenshot and URL, otherwise it sounds much like an AI hallucination to be honest. I do like the sloppy retro cover image though.
Would be nice to see. A screenshot would speak louder, doesn't need to be hosted yet. Best wishes.🎉🎉🎉
Fascinating
Wow! That's really fascinating - before the advent of AI, even if you'd use "no code" or drag & drop tools, you really wouldn't want to do this just on a phone, I think - AI does make this feasible ...
Can you disclose which AI tool you used? I somehow assume that simply ChatGPT or basic AI from Google or Meta won't cut it, but I could be wrong ... just curious, if you don't want to disclose it that's okay of course!
5 years ago, I read a post on (I think) Twitter, where John Carmack (yes the Doom guy), asked "Does anyone actually use a tablet for programming?" I read that and thought 'sure you can use DEX to atleast have a workspace, Jupyter for coding, but it's so impractical, I doubt anyone does it... Yeah... Like with Doom, he's always 10 steps ahead, the developer experience switching to Agent systems, means that quite literally, you can do it on a tablet or a phone, though compiling is still a hurdle, 2 seconds on a cloud server attached to a google drive wont hurt your wallet. Infact, I bet you pay more for your pc's electricity per month than you would just using a VM in the cloud with a tablet to connect. You never have to actually upgrade your system, if you need a burst, you pay for the 2 seconds you need a burst, if you want a LLM, you plug in the API, if you want a GPU, you rent a gpu, whatever size you want.
I think alot of us are probably coming to the realization that programmers dont need computers anymore. We never actually did really, we just thought we did, because that's how we had an IDE and a compiler... That's changed and maybe we should too? I havent really given it a try yet, but I think I will in the near future. With cloud compute as cheap as it is, might aswell?
"... programmers dont need computers anymore ..." - define "computer" :-)
Of course a smartphone is a computer, and a pretty powerful one - but, I "get" what you mean:
A "thin client" like a ChromeBook should be sufficient, if you do all your compiling/building/hosting "in the cloud", with a thin "web IDE" as the client ...
Such web client/cloud development environments have been around for quite some time, but AI coding might make this proposition a whole lot more practical and attractive ...
This is a really inspiring perspective on how AI is changing the way people create and express their ideas.
What impressed me most is that you focused not only on the tool itself, but on the importance of clarity, creativity, and knowing what you want to build. AI can accelerate the process, but the vision and human judgment behind the work are still the most valuable parts.
I also like your point about giving specific feedback instead of vague instructions. That is becoming an important skill when working with AI — the better we can communicate our ideas, the better results we can achieve.
Building a complete website from a phone is a great example of how technology is removing traditional barriers and allowing more people to turn their ideas into reality. Thank you for sharing your experience. It is a great reminder that the most important requirement is not always the perfect setup, but the willingness to start and keep improving.
What would be your ideal UX for doing this?