The Clock

Enter age, sex, ethnicity, location, then toggle lifestyle factors (smoking, exercise, …) and watch your expected lifespan clock move.

• frontend/ React + TypeScript + Vite. Dev server proxies /api to the backend.
• backend/ FastAPI. app/models/base.py is the LifespanModel interface; app/main.py serves whatever model it imports.

Run

# backend (http://127.0.0.1:8000)
cd backend
python3 -m venv .venv && .venv/bin/pip install -r requirements.txt
.venv/bin/uvicorn app.main:app --reload --port 8000
.venv/bin/pytest            # tests

# frontend (http://localhost:5173)
cd frontend
npm install
npm run dev

API

• GET /api/options – sexes, ethnicities, locations, lifestyle factors (the UI renders from this).
• POST /api/estimate – {age, sex, ethnicity, location, factors: {factorId: optionId}} → expected age, remaining years, baseline, and per-option delta_years (what switching to that option would change).

Contract lives in backend/app/schemas.py and frontend/src/types.ts; keep them in sync.

Plugging in real data

app/models/gompertz.py is a placeholder: Gompertz proportional-hazards baseline fed by approximate constants in app/data/placeholder.py (life expectancy by country/sex, hand-set hazard ratios, ethnicity HR = 1.0). The UI shows a "demo model" warning while engine.placeholder is true.

To use real datasets: implement LifespanModel (e.g. app/models/lifetable.py reading life tables + published hazard ratios from your DB), return EngineInfo(placeholder=False), and change the model import in app/main.py. The frontend needs no changes unless you add fields.

the-clock