Show HN: 0xCal – A calorie tracker where you just describe what you ate(apps.apple.com)
apps.apple.com
Show HN: 0xCal – A calorie tracker where you just describe what you ate
https://apps.apple.com/pl/app/0xcal-ai-calorie-tracker/id6749210009
3 comments
The natural language input is a nice touch. I always found traditional calorie trackers tedious with all the database searching. How does it handle portion sizes? Like if I say "had a burger" vs "had a big burger"?
Good question — this was one of the trickiest parts to get right.
When you say "had a burger," the AI makes an educated guess based on a typical burger and shows its assumption: "Burger (~200g patty, standard bun) — ~550 kcal"
If you say "had a big burger," it'll bump the estimate: "Large burger (~250g patty) — ~700 kcal"
The key is it always shows what it assumed — portion size, ingredients, etc. So you instantly see if it's off.
If it guessed wrong, you just correct it in the chat: "it was a smaller patty, maybe 150g" or "add cheese and bacon" — and it recalculates.
The goal was making corrections conversational instead of form-based. Feels faster than hunting through dropdowns.
Not perfect, but gets you in the right ballpark quickly. And if you want precision, you can always go detailed upfront: "200g beef patty, brioche bun, slice of cheddar, lettuce, tomato."
When you say "had a burger," the AI makes an educated guess based on a typical burger and shows its assumption: "Burger (~200g patty, standard bun) — ~550 kcal"
If you say "had a big burger," it'll bump the estimate: "Large burger (~250g patty) — ~700 kcal"
The key is it always shows what it assumed — portion size, ingredients, etc. So you instantly see if it's off.
If it guessed wrong, you just correct it in the chat: "it was a smaller patty, maybe 150g" or "add cheese and bacon" — and it recalculates.
The goal was making corrections conversational instead of form-based. Feels faster than hunting through dropdowns.
Not perfect, but gets you in the right ballpark quickly. And if you want precision, you can always go detailed upfront: "200g beef patty, brioche bun, slice of cheddar, lettuce, tomato."
Thanks! That's a really smart approach, showing the assumptions upfront makes it transparent.
Instead of searching databases, you just describe your meal: "scrambled eggs with toast and coffee" — AI calculates calories, carbs, protein, and fat. You can also snap a photo of food or a nutrition label.
Launched on Product Hunt yesterday and hit #5 with 200 upvotes — the feedback helped shape a lot of my thinking.
The accuracy scales with detail. "Had eggs" gives a ballpark. "2 eggs, 10g butter, 1 slice toast" gets precise. Grams = exact data.
I focused heavily on design — wanted something minimal that feels native to iOS. Most calorie apps look and feel like they haven't been updated since 2015.
Built with SwiftUI, syncs with Apple Health for personalized targets.
Would love feedback from the HN community on the approach. Happy to answer questions about the tech or design decisions.