MEALMARKOS  /  V1.0  /  WELLNESS INFRASTRUCTURE

One wellness platform to replace them all.

MealMarkOS is a multi-tenant platform that powers nutrition, strength training, and activity tracking for universities. Seeded with your dining data — open to every other meal. Branded for your institution. Owned by your students.

Already on a campus build? Jump to your school.

01The problemWhere wellness lives today

Wellness is scattered. Nutrition in one app, strength in another, activity in a third.

University wellness is a coordination problem disguised as a technology problem. The data exists — menus, allergens, activities, training logs — but none of it talks to itself, and none of it talks to the student.

TODAY

A scattered stack.

  • Campus menusBuried in a vendor portal nobody opens. Often a daily PDF.
  • Meal loggingGeneric consumer apps that don't know your dining hall.
  • Strength trainingA notebook, a Notes app, or a spreadsheet on someone's iPad.
  • Activity dataStuck in Strava, Garmin, Apple Health — never reconciled.
  • BrandA one-off app a department paid for in 2019. Stale by 2021.
ON MEALMARKOS

One timeline, your brand.

  • Campus menusYour menu feed normalized nightly into a queryable item database.
  • Meal loggingSearch the dining hall for real campus numbers — or log any meal by name or photo.
  • Strength trainingIron Logic accepts shorthand and parses it into structured sets.
  • Activity dataStrava sync, read-only, into the same timeline as meals.
  • BrandYour institution's identity. Optional custom domain.
02What's built inFour primitives

Four primitives. One timeline.

Every feature ties back to the same activity timeline so meals, lifts, and cardio are reconciled in one place — not five.

i.

Dining intelligence

Nightly automated scrapes of your menu and nutrition feed (NetNutrition and similar vendors), normalized into a queryable item database with allergens, micronutrients, and serving sizes.

Source
Your existing menu feed
Refresh
Nightly
Coverage
Allergens · micros · servings
ii.

Meal logging

Students search the dining hall they're standing in for your kitchen's exact numbers — then log everything else they eat by name (public nutrition databases) or by snapping a photo. One tracker, on-campus and off.

On campus
Live menu, per hall
Anywhere
Public nutrition DBs
Photo & label scan
Adversarial OCR
iii.

Iron Logic

An AI-augmented strength journal. Students brain-dump workouts in their own shorthand; specialized agents parse it into structured sets, build daily metrics (sleep, mood, caffeine), and answer coaching questions with cited reflections.

Input
Free-form shorthand
Output
Structured sets + metrics
Coach
Opt-in, cited answers
iv.

Connector ingest

Activity and recovery data flows in from the apps students already use — Apple Health, Strava, and more — landing on the same timeline as their meals and lifts. One picture of training, cardio, and nutrition instead of a tab per app.

Sources
Apple Health · Strava · +more
Brings in
Cardio · activity · recovery
Lands on
The unified timeline
03What students actually doThree flows · one Tuesday

From dining hall, to lift, to cited answer.

Every primitive ships against a real student workflow. Here's the same student, three taps deep, on the same Tuesday.

FLOW 01 · 6:42 PM

Log the meal that's actually in front of you.

Search the dining hall they're standing in for your kitchen's exact numbers — or log any off-campus meal by name or photo. Campus-accurate where it matters, universal everywhere else.

Source
Marketplace · live menu
Refresh
Nightly
Allergens
Carried through
Photo & label
Adversarial scan
MARKETPLACE · TUESDAY● OPEN
grilled chickenMarketplace ▾
  • Grilled chicken breast
    Entrée · 5 oz · contains: none
    220 kcal
    42P · 0C · 6F
  • Roasted sweet potato
    Side · 1 cup · contains: none
    180 kcal
    3P · 41C · 0F
  • Caesar salad, no croutons
    Salad · 1 bowl · contains: dairy, egg, fish
    160 kcal
    6P · 8C · 12F
Tray · Today560 kcal  ·  51P · 49C · 18F
FLOW 02 · 9:11 PM

Brain-dump in your own shorthand.

Iron Logic's Parser Agent reads the way lifters actually write — set-paren notation, RPE-per-set, BW pullups, ss supersets. It splits multi-day notes correctly or asks before guessing.

Input
Notes-app blob
Output
Sets · RPE · BW
Glossary
Per-user, learnable
On ambiguity
Ask, don't guess
BRAIN DUMP · RAW● DRAFT
Tuesday push day
bench 135x8, 185x5, 205x3 @8, 205x3 @9
ss DB OHP 35sx10/10/8 + cable tri 30x12,12,10
incline DB 50sx10x3
BW dips x12,10,8
felt strong, slept 7.5, 200mg caffeine
PARSED · GYM_SESSION● COMMITTED
Bench
135×8 · 185×5 · 205×3@8 · 205×3@9
DB OHP
superset
35×10 · 35×10 · 35×8
Cable tri
superset
30×12 · 12 · 10
Incline DB
50×10 × 3
Dips (BW)
12 · 10 · 8
sleep 7.5hcaffeine 200mgmood +1
FLOW 03 · WEDNESDAY AM

Ask the coach. It cites or admits.

The Coach Agent answers from the student's logged sessions and daily metrics. Every number cites a session date and set — and when it doesn't have the data, it says so instead of confabulating.

Retrieval
Reflections + metrics
Verifier
Post-hoc, every claim
Range
Loads on demand
Disclaimer
Not medical advice
IRON COACH · CHAT○ OPT-IN
How's my bench trending the last month?
IRON COACH

Up roughly 25 lb on your heavy triple — your top set went from 205×3 (Apr 02) to 225×3 (Apr 29). RPE on the top set stayed at @8 both sessions, so it's real progress, not just grindier work.

session · 2026-04-02session · 2026-04-15session · 2026-04-29
And my squat the same week?
IRON COACH

No squat sessions logged the week of Apr 22. I'd rather say that than guess. Want me to look at the prior month?

04Your wayTap or talk

Tap it. Or just talk.

Every feature works two ways. Want control? A fast, clean interface. Moving quick? Just say it. Most apps force one or the other — MealMark does both, on everything.

  • A real UI for hands-on logging & planning
  • Natural language for when you're moving fast
  • Switch anytime — same data, your choice
LOG A SET
Bench Press3 sets
225
lb × 5
225
lb × 5
225
lb × 5
05TrustOwned, branded, isolated

Institutions trust what they control.

Most campus wellness apps run every school off one shared database with a logo swapped in. MealMarkOS gives each institution its own isolated, branded environment — your data, your brand, your students.

A

Your data stays yours

Each institution runs in its own isolated environment. Student data is never pooled with other schools, never sold, and never used to train shared models. You own it — and you can export or delete it on request.

B

Branded as your own

Your colors, your logo, your name — optionally on your own domain. Students open your institution's wellness app, not a generic third-party product with your crest bolted on.

C

Built on your real menu

On campus, it runs on your actual dining feed, refreshed nightly — your kitchen's real nutrition, allergens, and serving sizes. Off campus, students log any meal from public nutrition databases or a photo.

D

Private by default

Students choose what they track; sensitive features stay off until they opt in. Access ties to your institution's single sign-on, so only your community gets in.

06Multi-tenantBranded per institution

One platform. Your institution's identity.

Every school runs on the same engine but wears its own brand — colors, logo, name, even your own domain. Students open your wellness app, not a generic product with a crest bolted on. The shell stays; the identity swaps.

ONE ENGINE · ISOLATED PER TENANT · LIVE ON ANY DOMAIN

TENANT · 01 OF 05LIVE
bowdoin.mealmarkos.com
bowdoin · bowdoin.mealmarkos.com · SSO
Menu feed
Per institution
Halls covered
Institution-wide
Iron Coach
Opt-in · cited
Email auth
Institution SSO
ONE PLATFORM · ANY INSTITUTION
07Current customersTenant directory

Pick your school.

Every tenant runs as its own app on its own surface. Students bookmark their school directly; nothing on the marketing site knows who they are.

Want your university here? hello@mealmarkos.com

08What you can askOne answer, your whole picture

One ask. It reasons across all of it.

Students type the way they think. The assistant connects their nutrition, training, sleep, and recovery at once — then answers with their own numbers, or acts on them. One picture, not five apps that never talk.

  • Why has my bench stalled the last three weeks?lifts · sleep · protein
  • Am I eating enough to support my training?meals · workouts · weight
  • What should I eat tonight to hit my protein goal?today's meals · goals
  • How is my sleep affecting my lifts?sleep · training
  • I'm cutting — am I holding onto muscle?weight · training · protein
  • Why has my bench stalled the last three weeks?lifts · sleep · protein
  • Am I eating enough to support my training?meals · workouts · weight
  • What should I eat tonight to hit my protein goal?today's meals · goals
  • How is my sleep affecting my lifts?sleep · training
  • I'm cutting — am I holding onto muscle?weight · training · protein
  • Plan my meals around leg day tomorrowroutine · macros
  • Did cutting caffeine actually improve my recovery?caffeine · sleep · mood
  • What's been sabotaging my weight loss?calories · activity · weight
  • Summarize how my week actually wenteverything
  • Why am I wiped out every Monday?sleep · weekend activity · meals
  • Plan my meals around leg day tomorrowroutine · macros
  • Did cutting caffeine actually improve my recovery?caffeine · sleep · mood
  • What's been sabotaging my weight loss?calories · activity · weight
  • Summarize how my week actually wenteverything
  • Why am I wiped out every Monday?sleep · weekend activity · meals

ONE ASSISTANT · NUTRITION · TRAINING · SLEEP · RECOVERY · ANSWERED WITH YOUR OWN NUMBERS

THE STACK/ RUNNING UNDERNEATH /
  • TENANT-ISOLATED
  • AURORA · POSTGRES
  • BEDROCK · CLAUDE SONNET 4.6
  • COHERE EMBED v3 · 1024-DIM
  • PER-TENANT BRANDING
  • SAML / SSO
  • WEARABLE & STRAVA SYNC
  • GROUNDED IN YOUR OWN DATA
  • NEXT.JS · REACT · TAILWIND
  • DOCKER · ECS FARGATE
  • TENANT-ISOLATED
  • AURORA · POSTGRES
  • BEDROCK · CLAUDE SONNET 4.6
  • COHERE EMBED v3 · 1024-DIM
  • PER-TENANT BRANDING
  • SAML / SSO
  • WEARABLE & STRAVA SYNC
  • GROUNDED IN YOUR OWN DATA
  • NEXT.JS · REACT · TAILWIND
  • DOCKER · ECS FARGATE
09TailoredHow we onboard

We scope it around how your campus runs.

We don't publish prices — university procurement runs on conversations, not pricing pages. We scope it, brand it, deploy it, and tune it as you grow.

01

Scope

We map your dining feed (NetNutrition or similar), your SSO, and which capabilities your campus needs — dining, training, signals, activity.

02

Brand

Your tenant is provisioned with its own schema and branded end-to-end: colors, logo, app name, and an optional custom domain. The same platform, your institution's identity.

03

Deploy

We launch to a pilot cohort first — showcase mode for stakeholders, then full access. SSO and email-domain gating switch on when you're ready to open it campus-wide.

04

Tune

A named owner keeps the menu feed healthy, adds capabilities as you grow, and tunes the deployment as your fleet of halls and programs evolves.

10FAQFor wellness & athletic buyers

Questions a buyer actually asks.

We've written these in plain language, the way a CSM would answer in a procurement call. If something else matters to your office, ask us.

  • Each tenant runs on isolated data — students at one institution never appear in another's queries, even at the database level. Authentication is gated to your email domain (e.g. only @duke.edu accounts can register on the Duke tenant). AI features ship behind an explicit per-student opt-in, with a plain-language disclaimer that the Iron Coach is not medical advice. We process data under a standard university DPA and can support data residency requirements on Enterprise.