# Building OopsCalorie: When Your AI Thinks Dinuguan Is Champorado 😂

> Source: <https://dev.to/chimkennuggyszdev/building-oopscalorie-when-your-ai-thinks-dinuguan-is-champorado-4961>
> Published: 2026-08-18 03:29:45+00:00

I’ve been building a side project called **OopsCalorie**, an AI-powered calorie and meal tracking app.

The idea sounded simple enough:

Simple, right?

Well...

Then we started testing it with Filipino food. 😂

One of the funniest parts of building OopsCalorie has been testing the food recognition.

At one point, our AI confidently looked at **dinuguan** and decided:

That's champorado.

Okay. I can *kind of* see where you were coming from. Both are dark, both can be served in a bowl...

But still. 😂

Then came **bagnet**.

AI:

Lumpiang Shanghai.

Bro.

Not even close. 😂

These bugs are funny, but they also exposed one of the more interesting engineering problems behind OopsCalorie:

**Image recognition is only the first step.**

Correctly identifying a meal — especially regional dishes — requires much more context than I initially expected.

When I started the project, I thought the difficult part would be estimating calories.

Turns out, before you can estimate:

**You need to know what the food actually is.**

And food can be surprisingly ambiguous from an image.

A photo might contain:

Even humans sometimes need context.

*"Is that pork adobo or humba?"*

*"Is that fried pork belly or bagnet?"*

Now imagine asking an AI to determine that from pixels alone.

This changed how I'm approaching the system.

Instead of treating the AI response as absolute truth, OopsCalorie is evolving toward a workflow where AI provides an intelligent estimate while the user still has the ability to provide context and correct it.

We're experimenting with things like:

**Image + user context**

A short description from the user can dramatically reduce ambiguity.

**Confidence scoring**

If the system isn't confident about what it's looking at, it shouldn't pretend that it is.

**Food candidates**

Instead of saying:

This is definitely X.

The system can consider multiple possible foods before making the final estimate.

**Filipino food awareness**

This is particularly important for us.

A lot of calorie trackers are built around foods that are easy to find in international nutrition databases.

But Filipino meals aren't always that straightforward.

Try estimating the calories of someone's homemade adobo where every household has a slightly different recipe. 😅

OopsCalorie is currently being built with:

The goal isn't to build another massive nutrition database with an AI label slapped on top.

I'm more interested in exploring how AI can make food logging feel less like data entry.

Take a photo. Give it some context if necessary. Get an estimate. Correct it when needed. Move on with your day.

OopsCalorie is currently in the **testing phase**.

And that's exactly why I'm sharing it now.

I want developers and early users to break it.

Upload weird meals.

Upload Filipino food.

Confuse the AI.

Find edge cases I haven't thought about.

And yes, if you manage to make it identify **lechon as a croissant**, please send me the screenshot. 😂

The Android and iOS versions are coming soon.

🌐 **Website**

[https://oopscalorie.com/](https://oopscalorie.com/)

🧪 **Become a beta tester**

[https://oopscalorie.com/beta-testers](https://oopscalorie.com/beta-testers)

📖 **Development stories & experiments**

[https://oopscalorie.com/blog](https://oopscalorie.com/blog)

I'm also open to collaborations with developers, designers, nutrition/fitness people, or anyone interested in experimenting with AI-powered consumer apps.

One thing this project keeps reminding me:

**AI can be incredibly impressive and incredibly stupid within the same five minutes.**

And honestly, that's part of what makes building with it fun.

Every ridiculous result becomes another edge case.

Every edge case becomes another improvement.

And every improvement gets us a little closer to making OopsCalorie actually useful.

I'll be sharing more about the technical decisions, AI experiments, failures, costs, architecture, and the occasional food-recognition disaster as development continues.

Until then...

**Dinuguan is not champorado.**

We're working on it. 😂