I tried using ChatGPT for meal prep. It looked great… until I started cooking.

TLDR:

I tried using ChatGPT for meal prep. The plans looked great, but the quantities were off, the shopping lists didn’t match the recipes, and I had to re-work the prompt and re-explain my preferences every time. So I built Bocalito. AI suggests the recipes, and regular code handles the quantities, shopping lists and preferences so they stay consistent. Would love your feedback if you give it a try

---

I know we're living in an era where everybody seems to be building apps. But I've been a software developer long before AI coding tools became a thing.

And one thing developers tend to do is build tools to make repetitive tasks easier. The problem is that when you already have a full-time job, you don't always have the time (or motivation) to build the tools you'd like to have for yourself.

How it all started

I started meal prepping a few months ago, thanks to my girlfriend, who introduced me to it. At first, I was skeptical. But once I got used to sitting down for lunch with everything already prepared, I realised how nice it was not to have to think about preparing or ordering food every single day. No more wondering what to eat tonight, checking whether I had all the ingredients, figuring out when to go shopping, or deciding whether to order food again. It was surprisingly liberating to have one less thing to think about.

I started looking at books and websites for meal plans, but they all had the same problem: the menus were already decided for me. They often included ingredients I didn't like, and I couldn't easily adjust recipes, quantities, or portions to match my preferences. It was not easy to prepare something I wanted to eat every day.

We're in the AI era, so why not just use prompts?

My first thought was: easy! I'll write a good prompt, give it to ChatGPT, Claude or another AI, and ask it to create a weekly meal plan. And honestly, the results looked amazing at first. Until I actually went into the kitchen and tried to follow one.

I'm not going to lie: it kind of worked. But there were some problems. Quantities could be wildly inaccurate, shopping lists didn't always match the recipes, and sometimes the suggested meals simply didn't make sense.

Every time I wanted to make a change, I had to go back to the prompt and explain my preferences all over again.

And when it comes to food allergies, I didn't feel comfortable relying on a generated answer without being able to properly check everything.

AI was a promising starting point, but it wasn't quite the solution I needed.

Why not build a proper app instead of another AI wrapper?

Remember, I'm a developer.

So I thought: why not build something that combines the flexibility of AI with the reliability of actual software? AI is great at generating ideas and suggesting recipes. But software is much better at handling certain things consistently: calculating quantities, checking shopping lists against recipes, organising cooking steps, remembering preferences and making sure changes are properly reflected throughout the plan.

That's how Bocalito came to life.

I use it every week for my own meal prep, and whenever I can, I keep improving it. Eventually, I decided to make it available to anyone who might be facing the same problem — or simply wants to organise their meals for the week without spending hours planning everything.

I'd love to hear your feedback

If you have a few minutes, I'd really appreciate it if you gave Bocalito a try.

I'm particularly interested in hearing:

• Does the meal-planning process make sense?

• Are the menus and recipes actually useful in real life?

• What's missing that would make you want to use it every week?

And if you know someone who does meal prep, feel free to send it their way! I'm building this for a problem I personally have, but I'd love to make it useful for other people too.

Give it a try: Bocalito | Meal prep in minutes