Recently, I started experimenting with AI tools like Cursor to see how they could extend an engineer’s capabilities — not to replace the engineer, but to amplify what one person or company can achieve with the right tooling.

My first project? Building a personal website. I chose this for three reasons:

I needed a personal website.
I wanted to start small with test-driving AI-Assisted Engineering.
Being recently laid off gave me the bandwidth to do this quickly.

For someone who hasn’t written production code in years, this project could’ve easily taken me weeks. But with an AI assistant translating my ideas into working code, I was able to build a fully custom, full-stack site over a few scattered two-hour sessions within 14 days.

Using only prompts and no manual coding, AI helped me:

Scaffold a complete frontend.
Build (and later refactor) a CMS backend using Postgres.
Generate a GraphQL API to connect the frontend and backend.
Set up pre-commit hooks for code quality.
Configure a CI pipeline to validate and deploy each app independently.

The experience was surprisingly fluid. It felt like having a team of engineers - I fed the agents my vision and stepped in only to guide or correct when needed. The result is a functioning website with backend tools to help me produce content that’s been stuck in my head for years.

To those asking “how does the code look?” or “how fast does it run?” - This wasn’t about perfect architecture or production-grade performance. The goal was to go from an idea to something real I could share. Sure, I could’ve reached for an off-the-shelf CMS or spent more time refining the system, but that wasn’t the point. The real value was in how fast I was able to move, and how satisfying it was when AI got it right.

Now that my site is up and running, I’m already thinking about the next AI-assisted experiment. There’s a lot more to uncover, and I’m excited to keep learning.