FinGuard AI

Inspiration

Managing finances across multiple banking apps was exhausting and risky. We kept losing track of charges, forgetting subscriptions, and worrying whether suspicious transactions were actually fraud. We knew AI could solve this problem by actively protecting people's money, rather than just tracking it, so we set out to build a financial guardian that works 24/7.

What it does

FinGuard AI connects to your bank accounts and uses Google Gemini to analyze every transaction in real-time for fraud patterns and suspicious activity. It automatically detects forgotten subscriptions draining your wallet and assigns each transaction a risk score to catch threats before they become losses. The built-in AI chatbot helps you understand your spending habits, improve your FinGuard Score, and get personalized advice to boost your financial wellbeing.

How we built it

We built FinGuard AI using Next.js and Tailwind for a fast, responsive web experience, with Clerk handling secure authentication. The backend utilizes Plaid's API to retrieve the user's data from their bank accounts. Google's Gemini API was used to create smart and decisive decisions for risk analysis and fraud detection while offering a great way to learn more through the chatbot.

Challenges we ran into

There were many challenges since Plaid's API is a bit tricky to implement. Although it is not 100% functional at this time, we utilize mock data to show that our AI and our algorithm work well to detect risky or fraudulent behavior.

Accomplishments that we're proud of

We're proud of building a production-ready fintech app with bank-grade security in just a hackathon timeframe. The AI chatbot feels genuinely helpful rather than gimmicky, and our clean UI makes complex financial data actually understandable. We were proud to accomplish so much in a tight span of time.

What we learned

We learned that AI should augment, not replace, traditional rule-based systems. Most importantly, we discovered that users don't just want budgeting tools; they want protection and peace of mind about their financial security. We realized how helpful and pivotal AI is to the banking security landscape.

What's next for FinGuard AI

We plan to add predictive fraud prevention that warns users before suspicious merchants even charge their cards. Expanding our recurring charge detection to automatically negotiate lower bills and cancel unwanted subscriptions would save users real money. We also want to integrate with more financial institutions beyond Plaid and build mobile apps so users can get instant fraud alerts wherever they are.