Out of all my side projects, this is the one I take most seriously — not because it’s the smartest idea in the world, but because I’ve put enough time into it that walking away now would feel like a waste.
Somewhere in your 30s, you start getting invited to a steady stream of “adult” occasions: weddings, birthdays, housewarmings, becoming someone’s godparent, hosting the office gift exchange. Where I’m from, gift registries and wishlists aren’t really part of the culture. And while any gift is appreciated, I kept wondering how much easier it would be if people could give and receive gifts without the slightly awkward “So… what do you actually want?” conversation.
Maybe that’s just the introvert in me talking. Less small talk, more “here’s the exact thing that will make me unreasonably happy.”
Picking a Name
My first instinct was wishlist.co — a nice play on “Wishlist Ko” (Filipino for “my wishlist”). Catchy, and in my head, it was already The Name.
I learned the hard way that building a brand should start with securing the domain, not falling in love with a name you don’t own yet. Turns out wishlist.com already exists in the same space, so wishlist.co would have meant constant brand confusion. And when I checked, the domain itself was being squatted for roughly ₱4 million — a price that made the decision very easy.
So, wishlog.co it is. What started as a backup has become its own thing entirely, and I’ve genuinely grown to like it more.
What I Want wishlog.co to Be
Plenty of sites already do some version of this — but most are narrowly scoped, and nearly all of them ask you to create an account, choose a password you’ll forget by next month, and stay locked inside their ecosystem.
I wanted something simpler: one place that works as your gift registry, your personal wishlist, and your tool for hosting gift exchanges — all under one link.
Birthdays, weddings, housewarmings, holiday exchanges — same concept, same shareable link, less friction.
Building With AI as a Not-Quite-Developer
Here’s the part that made this project genuinely hard: I’m not a full-time developer. I’ve been comfortable with HTML and CSS since the Friendster era, but wiring together a working back end — the part where nothing is allowed to quietly break — is a different skill entirely.
That’s where AI came in, and it’s been more useful than I expected — though nowhere near as effortless as “type one prompt, get a finished website.” It took thousands of prompts, a fair amount of troubleshooting, and more than one moment where the AI confidently rewrote code I never asked it to touch. At one point, ChatGPT actually suggested I take a break, which — given how the conversation was going — was fair advice.
A few things I picked up along the way:
Every model has its own personality. Gemini tends to move fast and keep its code simple, while ChatGPT is more thorough — sometimes to a fault, producing longer, more heavily commented output.1 Taming that output taught me a lot about refactoring.
Specificity is non-negotiable. Left unchecked, AI will happily rewrite an entire file when you only wanted one function changed — the same “almost right, but not quite” pattern that’s now the single most common frustration developers report with AI coding tools.2 “Please only touch X, Y, and Z” became one of my most-used phrases.
Organization matters more than it seems like it should. Tracking changes, keeping backups, and naming files sensibly turned out to be essential — a proper local testing environment would have saved me a lot of grief. A “simple” wishlist tool looks straightforward from the outside; underneath, it’s a surprising amount of logic and edge cases.
On AI and the Work We Do
There’s more to say here — both about the technical side of this project and what it says about how I work — but that’s a topic for another post.
For now, my takeaway is this: AI is a genuinely powerful tool when used thoughtfully. Working in the digital space, it’s changed how I approach problems day to day. Whether it eventually reshapes my job is an open question, but the more useful mindset isn’t resistance — it’s figuring out how to work alongside it. That lines up with what a lot of current research on developers and AI is finding too: the tools are augmenting people’s work far more than replacing it outright.3 Treat it as a capable but occasionally chaotic collaborator that does its best work with clear direction.
Try wishlog.co
I’d love for you to give it a try. It isn’t 100% bug-free — nothing ever is — but it’s usable today. Create a wishlist, set up a registry, or host a gift exchange, and see how it feels.
There’s still work ahead (including the occasional email landing in spam — if you spot one, marking it “Not spam” genuinely helps). But what started as just another side project idea is now something real that people can use, and that alone feels like a win.
Sources:
1 DataCamp. (2025, November 25). Gemini vs. ChatGPT: Which AI model performs better? https://www.datacamp.com/blog/gemini-vs-chatgpt
2 Stack Overflow. (2025). AI section, 2025 Stack Overflow Developer Survey. https://survey.stackoverflow.co/2025/ai/
3 Houck, B., Lowdermilk, T., Beyer, C., Clarke, S., & Hanrahan, B. (2025). The SPACE of AI: Real-world lessons on AI’s impact on developers. arXiv. https://arxiv.org/pdf/2508.00178