FinHealth AI – AI Financial Health Monitor

Inspiration

Most people have income and expenses, but few truly understand them. For Maestro Hack Day, I built an AI-powered financial health monitor that turns raw numbers into actionable advice. The idea: combine Fintech, IoT, and AI in one tool—a coach, not a spreadsheet.

How I Built It

FinHealth AI is a Next.js 14 app: dashboard with KPIs, charts, spending by category, goals, upcoming bills, plus an AI analysis block (Google Gemini) and Stripe for Pro subscriptions. The savings rate formula:

$$\text{Savings Rate} = \frac{\text{Income} - \text{Expenses}}{\text{Income}} \times 100\%$$

Stack: Next.js 14, Tailwind, Recharts, Gemini 2.0 Flash, Stripe.

What I Learned

• How to integrate Gemini and craft prompts for structured financial reports
  
• Lazy-loading Stripe in API routes to avoid build errors when keys are missing
  
• Balancing information density with a clean, readable dashboard UX
  

Challenges Faced

• Gemini 404: Switched from gemini-1.5-flash to gemini-2.0-flash
  
• Stripe build: Resolved via dynamic imports in the checkout route
  
• UX overload: Simplified layout and hierarchy so key info stands out
  

Built With

• 14
• api
• css
• gemini
• google
• icons
• lucide
• next.js
• react
• recharts
• stripe
• tailwind
• typescript