I’ll also talk about what we tried for: marketing, bootstrapping a part-time project, and more. Follow along if you’re curious about building and launching a product 🚀

We’re thrilled to have launched the app, but the journey doesn’t stop here. In the coming days, I’ll share more about:

- The tech stack and challenges we faced
- How we designed features and tackled AI quirks.

Over time, those challenges turned into new features. We’ve iterated, refined, and finally launched the app!

You can try it for free for 14 days here: https://www.menumagic.ai/landing

We started to think this app could help others too. But there were still challenges:

Splitting up the shopping list had its own quirks.
Reviewing the AI-generated plans sometimes felt tedious.
Each problem became a new feature—or a bullet point on our roadmap.

I dug up my old project, revamped it with AI, and convinced Monica to give it another shot. Suddenly, meal planning became a 5-minute Friday ritual. We’d tweak the plan, split the shopping list, and hit the store. We were our first users—and it was working!

Fast forward a couple of years: ChatGPT and AI became all the rage. After tinkering with it, I realized it could generate meal plans that fit our needs without requiring tons of input.

It was a game-changer.

Unfortunately, the app brought new frustrations:

Adding recipes was a pain.
Inputting ingredients for the shopping list was tedious.
Eventually, I abandoned the project, and it ended up in my 'graveyard' folder 😞.

At the time, I was a mobile developer in Turin. So I cobbled together an app to mix recipes, minimize repetition, and generate a shopping list. It took a few weekends, but it was somewhat usable in a few weeks.

We knew we needed to make this chore… less of a chore. We tried a bunch of meal-planning apps, but they were mostly focused on trying new recipes. That wasn’t us—we wanted to plan (mostly) meals we already knew how to make.

But as much as I hated sitting down after work to plan meals for the week, it saved us a ton of time. We were both busy with work and studies, and each trip to the supermarket meant:
🚗 40 minutes of driving
🛒 Time deciding what to cook
⏳ Endless checkout lines.

Monica was used to planning meals weekly—it was something her mom always did. But when we moved in together, I quickly realized I wasn’t contributing much to the process. And honestly? Meal planning felt like a massive chore.

Checks out lmao

It's the same plan and shopping list you'd eventually make yourself.

But it'll take you 1 min instead of ~40 (on average) every week.
And they're shared and synced, so you can split up at the supermarket and speed up that chore as well.

1-minute weekly meal plan and shopping list, synced in realtime with your family
@menumagic-ai.bsky.social

Features hardly ever work in isolation, so adding on/off switches actually means splitting logic in all the places affected by that feature, for every feature flag. It can quickly become overwhelming and hard to track.

In theory this can be planned for ofc, but it's often not worth the overhead

The best time to start looking for your wallet is only *after* you've been told how much you're supposed to pay 😬