The day at a glance
A calorie ring, macro breakdown and fiber tracking — the whole day readable in the second before you put the phone down.
FitMeSoon
Tell it what you ate. It does the rest.
A nutrition platform built on roughly two million foods from the USDA database — where logging a meal is a barcode, a photo, or a sentence, and a coach that actually knows your week decides what tomorrow should look like.
In active development · iPhone and Apple Watch
~2M
Foods from USDA FoodData Central
70M+
Individual nutrient values ingested
3
Ways to log a meal
2
Devices — iPhone and Apple Watch
Why another calorie app
People quit tracking because tracking is tedious.
Searching a database, guessing a portion, correcting a wrong entry — four times a day, every day. Most people last a fortnight. FitMeSoon is built around removing that friction rather than adding more charts on top of it.
Logging
Three ways in. All of them fast.
Barcode, photo, or words
Whichever is quickest at the time.
Scan the packet. Photograph the plate. Or just type “two eggs, toast and a flat white” and let the app work it out. Each route ends in the same place — items on your day with real nutrition attached, ready for you to correct before saving.
- Barcode scanning for anything packaged
- Photo logging for meals that never had a barcode
- Natural language for when typing is simply faster
- Missing product? A submission flow lets you add it, so the next person finds it
Scan a barcodeStraight to the exact product
Photograph the plateRecognised, portioned, itemised
“chicken salad, no dressing”Parsed into logged items
Not in the catalogue?Submit it — everyone benefits
Every route ends in a reviewable entry, never a silent guess.
The data
Two million foods, curated — not scraped.
The catalogue comes from USDA FoodData Central, the United States Department of Agriculture's public food composition database. A streaming pipeline ingests roughly two million foods and more than seventy million individual nutrient values.
Raw government data is not a product, though. A curation and de-duplication pass turns it into something searchable — so “chicken breast” returns one sensible answer rather than four hundred near-identical rows.
USDA FoodData CentralPublic, authoritative composition data~2M
Nutrient valuesStreamed and indexed70M+
Curation & dedupOne good answer, not four hundred✓
Exercise video libraryBundled alongside the food catalogue✓
A real ingestion pipeline behind a simple search box.
The coach
Advice that has read your week.
A generic “eat 1,800 calories” is not coaching. FitMeSoon's coach works from what you actually logged — the days you overshot, the protein you keep missing, the training you skipped — and adjusts from there.
It answers in cards you can act on, and there's a voice mode for when you'd rather ask a question than tap through a screen.
Adaptive guidanceBuilt from your real logged history
Actionable cardsSpecific, not motivational filler
Voice modeAsk out loud, hands busy
Generated workoutsWith a video for every exercise
Coaching that changes when you do.
The rest of it
A whole platform, not a single screen.
On your wrist
A companion Apple Watch app, so progress and logging travel with you instead of living in your pocket.
History that means something
Day summaries and calorie history let you edit the past honestly and see the trend rather than one bad Tuesday.
Onboarding that asks first
A guided flow through your age, body, activity level, diet and goal — so the first number you see is yours, not a default.
Exercise video library
Every generated workout comes with video for the movements, cached on the device so a gym with no signal still works.
Real infrastructure
A Python and FastAPI backend on PostgreSQL with Redis, containerised and deployed with Docker, plus an admin panel behind it.
In active development
FitMeSoon is not on the App Store yet. The backend, the iOS and watchOS apps, the USDA ingestion pipeline and the admin platform are all built and being refined toward a release. If you'd like to hear when it ships, get in touch — no list, no marketing emails, just a message when it's out.
Specifications
The details.
Platform
iOS and watchOS — iPhone app with an Apple Watch companion
Built with
Swift and SwiftUI on the client; Python and FastAPI on the server
Data source
USDA FoodData Central — roughly 2 million foods and 70M+ nutrient values, curated and de-duplicated
Food logging
Barcode scanning · photo recognition · natural-language entry · manual search, with a submission flow for missing products
Coaching
Adaptive AI coach with card-based guidance and a voice mode, working from your logged history
Training
Generated workouts with an exercise picker and an on-device-cached video library
Tracking
Calorie ring, macros, fiber, body stats, goals, day summaries and history
Infrastructure
PostgreSQL with Alembic migrations, Redis, Docker and Caddy, plus a web admin panel
Status
In active development — not yet released on the App Store
Questions
Good to know.
Is FitMeSoon available yet?
Not yet — it's in active development and hasn't been released on the App Store.
Get in touch and you'll hear when it launches.
Where does the nutrition data come from?
USDA FoodData Central, the United States Department of Agriculture's public food composition
database. A streaming pipeline ingests roughly two million foods and more than seventy million
individual nutrient values, then curates and de-duplicates them into something you can
actually search.
How does AI food logging work?
Rather than searching a database and entering portions by hand, you describe a meal in a plain
sentence or photograph it, and the app turns that into logged items with nutrition attached.
You always see the result before it's saved, so a wrong guess is a quick correction rather than
a wrong day.
Is there an Apple Watch app?
Yes — a companion watchOS app, so the day's progress is on your wrist rather than only in your
pocket.
What if a barcode isn't recognised?
You're not left at a dead end. A submission flow lets you add the missing product, so the
catalogue improves for the next person who scans it.
Coming to iPhone and Apple Watch.
Be the first to know when FitMeSoon ships.
More from Babak
