Vibe-coding our wedding website – Jan-Lukas Else


AI generated summary: I’m vibe-coding my wedding website, strengthening my Angular skills with AI’s help. Copilot with Opus is speeding up development, allowing me to focus on refining rather than boilerplate. It’s making the process fun, even with a custom Go backend and CI/CD pipelines!

For our wedding, I planned to develop a small site where guests can get more information, confirm their attendance, and view and upload photos. I started the base for that website in August and also bought a domain a few months ago. But I didn’t finish the coding. The design shouldn’t be as simple as my blog’s design, and I chose Angular for the frontend, as I also wanted to strengthen my Angular skills for professional reasons. Later, I wanted to build the backend using some n8n workflows.

The professional need to elevate my Angular skills decreased, so I procrastinated and developed new features for my blog instead.

Until a few days ago. With Anthropic releasing Opus 4.6 just recently, and GitHub Copilot already supporting it (and of course, me having a GitHub Copilot Pro license via the GitHub Education Pack) I found new motivation to continue my vibe coding on this site.

Opus is quite good at trying to figure out things itself. Tell it my requirements, and it builds it. Tell it what I want different, and it modifies it. Within just a few hours I have something pretty usable, which would have taken much longer to set up manually. Sure, it’s not perfect, and I need to touch a few places myself, but I think just correcting what’s not how I want it to be is much more fun than building all the boilerplate myself.

I cancelled my plans to build the backend with n8n and opted for a “simple” Go-based backend instead. Sometimes code is just better than connecting blocks. Now I have Angular on the frontend, and the backend sends mails on attendance confirmations and loads photos for a gallery from an S3-compatible object storage bucket. In addition, it now also utilizes (a self-hosted instance of) imgproxy to optimize the images on the fly (and caches the optimized images on S3 again).

Copilot with Opus also set me up with all the required pipelines (GitHub Actions) to test and build things, and even created a nice helper script for local development.

At work I’m working non-AI-assisted, so with AI, building projects like this doesn’t feel like work anymore. 😄

Of course, I could have just used something like Weddybird, which wouldn’t have made any difference to the cost of the wedding, but I’m also known throughout the family as the tech guy or developer, so I don’t want to disappoint anyone’s expectations. It’s also fun, and despite all the AI, I’m learning something in the process.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *