A few years ago, building a personal website from scratch would have meant crossing into an entirely different professional stack.
Not because I couldn't figure it out. I've spent years building data pipelines, managing ETL processes, and navigating complex enterprise systems. I know how to learn things.
The barrier was activation energy — the cost of context-switching across too many disciplines at once:
- HTML and CSS
- Git and GitHub
- Hosting and deployment
- Payment integration
- Domain configuration
- Accessibility standards
Each of those represents its own tooling, its own failure modes, its own learning curve. In any given week, there are already enough of those in my actual work.
This week, I built my website — camrychapman.com — in about a day and a half.
And the most surprising part wasn't the speed.
It was how different the process felt.
AI Didn't Replace the Work — It Changed the Nature of It
The website is completely built with HTML and CSS. No website builder. No WordPress. No drag-and-drop templates.
I learned Git and GitHub while building it. The site is hosted on GitHub Pages. I integrated Stripe payments. I configured a custom domain. And the total direct cost was about $7 for the domain name.
But the real story is how AI transformed the process.
I used AI constantly throughout development — especially Claude — not as a replacement for thinking, but as a collaborative acceleration layer.
AI helped:
- explain concepts quickly
- troubleshoot styling issues
- generate and refine code
- review accessibility concerns
- improve responsiveness
- structure commits
- handle Git pushes
- reduce context-switching
- shorten debugging cycles dramatically
What would normally require searching Stack Overflow threads, watching tutorials, reading documentation, and hours of trial-and-error often became a short conversation.
That changes everything.
The Biggest Shift: Momentum
The most powerful thing AI gave me was momentum.
Normally, beginner friction kills progress.
You hit an error. You lose an hour. You get discouraged. You stop building.
But with AI, the feedback loop became immediate.
I could ask:
- "Why is this flexbox layout breaking?"
- "Why does this page not match my other pages?"
- "How do I push this to GitHub?"
- "How do I integrate Stripe?"
- "Can you review this for accessibility issues?"
And instead of getting stuck for hours, I could continue moving.
That momentum compounds.
A Single Person Can Now Operate Like a Team
What struck me most: AI compresses the gap between idea and execution.
During this project, I was effectively acting as designer, developer, project manager, content writer, deployment engineer, and QA tester — not because I mastered all those disciplines overnight, but because AI lowered the activation energy required to operate across them.
We talk a lot about AI replacing jobs. But one of the more immediate shifts is simpler: AI expands individual capability. A single motivated person can now execute projects that previously required a team.
That has real implications for how organizations should think about staffing, capacity, and what "expertise" actually means.
Accessibility Was One of the Most Interesting Parts
One thing I appreciated was using AI for accessibility reviews.
AI caught issues I would not have thought about:
- color contrast problems
- semantic structure concerns
- navigation consistency
- responsive layout issues
- accessibility labeling
That felt meaningful because accessibility is often treated as an afterthought, especially in small personal projects. Instead, it became part of the workflow from the beginning.
GitHub in Context
Git and GitHub have always made sense to me conceptually. Version control is a familiar idea in data work — you don't overwrite source tables without a paper trail.
What AI changed wasn't the understanding. It was the translation from concept to command, in context, without breaking flow. Instead of pausing to search documentation or debug cryptic error messages alone, I could ask what went wrong and immediately continue.
That kept the momentum going in a way that documentation alone rarely does. And more importantly, learning became embedded inside the workflow itself — which is a much more effective way to retain it.
The Real Efficiency Gain Isn't Automation
The biggest misconception about AI is that efficiency only comes from automating tasks.
What I experienced was different. The efficiency came from:
- reducing uncertainty
- shortening research cycles
- maintaining momentum
- lowering cognitive overhead
- accelerating learning
- reducing context switching
That's a very different type of productivity.
AI didn't "build the website for me." It helped remove the invisible friction that usually slows people down.
This Was the Best AI Integration Experience I've Had
I've experimented with AI in many contexts: writing, analytics, coding assistance, brainstorming, research.
But this was the first time where it truly felt integrated into the workflow instead of sitting beside it.
The interaction became collaborative:
- I would guide the direction.
- AI would accelerate execution.
- I would refine the result.
- AI would help iterate faster.
That loop felt incredibly natural.
Final Thoughts
The website is just the artifact.
What it represents is more interesting: a meaningful drop in the activation energy required to execute across disciplines. In less than two days, with minimal direct cost, I built and launched a fully functional personal website with custom styling, responsive pages, Stripe integration, GitHub hosting, a custom domain, accessibility improvements, and version control workflows.
AI didn't eliminate the need for effort. I still had to make decisions, evaluate outputs, refine designs, and persist through problems. But it dramatically increased the speed at which I could learn, iterate, and execute.
For anyone working in data, leadership, or institutional strategy: the question isn't whether AI will change how your teams work. That's already happening. The more useful question is what becomes possible when individuals can operate across domains they couldn't before — and what that means for how you build teams, scope projects, and define expertise.
I don't have all those answers yet. But building this site gave me a much clearer sense of the right questions to ask.