AI-Powered Personal Finance Intelligence Platform

Transaction history with automatic categorization and recurring expense detection
Dashboard showing real-time balances, budget usage, and monthly spending insights
Mobile-responsive landing view showcasing AI-driven financial automation and real-world use cases.
Automated monthly financial report with budget summary, spending breakdown, and AI-generated insights.

What it does

Finexa automates personal finance management instead of relying on manual tracking. Users can upload receipts, record transactions, and set budgets, while the system automatically categorizes expenses, detects recurring spending, and monitors budget usage. When spending crosses safe thresholds, users receive alerts, and at the end of each month they receive AI-generated reports with clear summaries and suggestions. A simple dashboard provides a quick view of balances, trends, and monthly spending without overwhelming the user. RAHH Factor:Finexa runs in beast mode through event-driven automation, continuously monitoring spending, enforcing budget limits, and delivering AI-powered alerts and reports without requiring constant user interaction,This helps users stay informed about their spending, gain clear awareness of financial habits, and receive AI-driven suggestions to make better decisions.

How I built it

The platform is built using Next.js with a PostgreSQL database managed through Prisma and Neon DB to ensure reliable data modeling and transactional integrity. Authentication and user isolation are handled using Clerk. Core automation, including budget alerts, recurring transaction processing, and monthly report generation, is powered by Inngest background workflows. Google Gemini AI is used for receipt understanding and financial insight generation, while Arcjet enforces rate limiting and bot protection to prevent API abuse and ensure system stability at scale.

Challenges We ran into

A key challenge was designing automation that feels helpful rather than noisy. Alerts triggered too frequently reduce trust, while delayed alerts remove value. Another challenge was safely handling AI-generated receipt data by validating and structuring outputs so financial records remain consistent. Coordinating background workflows with real-time user actions and implementing effective rate limiting without impacting user experience required careful system design.

Accomplishments that I am proud of

We am proud of building a system that continues to deliver value even when users are inactive. Automated reports, proactive budget enforcement, and event-driven workflows make Finexa useful without constant attention. Implementing abuse prevention, secure authentication, and background automation gives the platform real-world readiness beyond a typical hackathon demo.

What we learned

We learned that financial applications are built on trust, predictability, and reliability rather than feature count. Automation must be intentional, AI outputs must be constrained, and security cannot be optional. Building background workflows and handling edge cases highlighted how production systems differ from simple CRUD applications.

What’s Next

Finexa is evolving from post-spending analysis into a pre-purchase decision engine. Planned extensions include budget-aware price monitoring that alerts users when products hit target prices, deeper spending intelligence, long-term forecasting, and automation that actively controls financial outcomes instead of reacting after money is spent.