Gamified financial literacy platform for immigrants. Duolingo-style learning focused on US personal finance (banking, credit, taxes, investing, retirement, immigration finance).

• Modules: Banking, credit, taxes, investing, retirement, insurance, major purchases, consumer protection, crypto, financial planning
• Lesson flow: Content blocks (text, quiz, interactive), mastery-based unlock, linear progression within sections
• Gamification: Hearts (lives), gems (currency), XP, daily streaks, quests (daily/weekly/special), leaderboards
• Auth: Google OAuth, guest/demo mode

• BKT (Bayesian Knowledge Tracing): Tracks skill mastery per knowledge component
• FSRS: Spaced repetition scheduling for review timing
• IRT (Item Response Theory): 2PL model for item difficulty calibration
• Content selector: Combines BKT, FSRS, IRT with Thompson Sampling. Targets p(correct) ~0.6 (zone of proximal development)
• Recommendations: Next lesson/section based on mastery, due reviews, and content selector output

• Course prioritization: Profile (country, visa, goals, experience) + diagnostic test results drive course order
• LLM personalization: Cultural bridge text, visa-specific content, simplified explanations by English level, hints, wrong-answer explanations
• Diagnostic test: Optional placement; informs domain_priority and domain_mastery for recommendations

• Backend: Flask, MongoDB, gunicorn
• Frontend: React 18, TypeScript, Vite, Tailwind, Zustand, Framer Motion
• LLM: Gemini (OpenAI fallback)
• Voice: ElevenLabs TTS + transcription; Google TTS fallback
• Payments: Stripe
• Deploy: Railway (backend), Vercel (frontend)

backend/
  app.py              # Flask app entry
  auth.py             # OAuth
  blueprints/         # curriculum, adaptive, learners, quests, streaks, leaderboard, chat, social, payments, translate
  services/           # learning_engine, bkt, scheduler (FSRS), irt, content_selector, personalization, voice
  database.py
  mongo_collections.py
frontend/
  src/
    pages/            # Learn, Lesson, Section, Profile, Shop, Quests, Leaderboard, etc.
    components/
    services/api.ts   # API client
    stores/userStore.ts
    contexts/

Backend

cd backend
pip install -r requirements.txt
cp .env.example .env  # add MONGO_URI, GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, etc.
# or use requirements-prod.txt for lighter deps (no torch)
gunicorn app:app --bind 0.0.0.0:5000

Frontend

cd frontend
npm install
# VITE_API_BASE_URL=http://localhost:5000 in .env
npm run dev

Backend: MONGO_URI, GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, FLASK_SECRET_KEY, FRONTEND_URL, GEMINI_API_KEY (or OPENAI_API_KEY), ELEVENLABS_API_KEY (optional), STRIPE_SECRET_KEY, STRIPE_WEBHOOK_SECRET.

Frontend: VITE_API_BASE_URL, VITE_GOOGLE_CLIENT_ID, VITE_STRIPE_PUBLISHABLE_KEY.