-
-
User Registration interface for Trace, enabling secure account creation to access the payment verification ,subscription tracking dashboard.
-
Trace's secure Login interface, allowing users to authenticate and access their personalized payment verification dashboard.
-
Trace's personalized Home dashboard, greeting the authenticated user with quick access to core verification and subscription tracking tools.
-
Trace's Verify Claim in action: 95% confidence match with clear bank statement evidence.
-
Trace's Subscription Leak scanner detecting 7 recurring charges totaling $93.96/month, with options to download dispute evidence.
-
Trace's Trap Scan detecting a price jump trap on a Google One subscription, exposing hidden billing increases with clear evidence.
Inspiration
Freelancers and small businesses often face financial uncertainty. A client may claim that a payment has been sent, but verifying whether the money actually arrived can be difficult. At the same time, recurring subscriptions, forgotten free trials, and unexpected price increases can quietly affect their finances.
We wanted to build a platform that helps people understand their transactions, verify payment claims with evidence, and identify recurring charges that deserve attention.
This inspired us to build Trace — a smarter way to bring clarity and transparency to everyday business finances.
What it does
Trace is an AI-powered financial clarity platform designed for freelancers and small businesses.
Its key features include:
AI-Powered Payment Claim Verification: Users can enter a payment claim in natural language. AI extracts important details such as the client name, amount, and claimed date. Trace then compares these details with transactions in the selected bank statement and provides a verification result with a confidence score and matching evidence.
Subscription Analysis: Trace analyzes transaction history to identify recurring payments and organize subscription-related information, helping users understand their regular expenses.
Trap Scan: Trace looks for potentially concerning transaction patterns, including sudden price increases and small initial charges followed by larger charges. It presents the evidence so users can investigate further.
PayPal Integration: Trace integrates with the PayPal Developer Platform through its Sandbox environment to support PayPal-powered invoice workflows.
Together, these features help users make more informed decisions about their payments and recurring expenses.
How we built it
We built Trace using a full-stack architecture.
Frontend:
- React
- Vite
- Tailwind CSS
Backend:
- Node.js
- Express.js
- MongoDB
AI Integration:
- Groq / Gemini AI for understanding natural-language payment claims and extracting structured payment details.
Payment Integration:
- PayPal Developer Platform Sandbox for invoice-related workflows.
The application processes bank statement transaction data, extracts payment information from user claims using AI, and compares the extracted details against the selected statement.
For our prototype demonstration, we use fictional sample bank statement data that has already been loaded into the application.
Challenges we ran into
One of our biggest challenges was making payment claim verification reliable. Natural-language claims can contain incomplete or differently formatted information, so extracting the correct client name, amount, and date was important.
Another challenge was working with bank statement data, where transaction descriptions and formats can vary. We also had to identify recurring transaction patterns without incorrectly treating every unusual charge as fraud.
Integrating PayPal Sandbox into the application was another important technical challenge.
Throughout development, we focused on making the results understandable and presenting transaction evidence rather than simply displaying a yes-or-no answer.
Accomplishments that we're proud of
We are proud to have built a working full-stack prototype that brings several useful financial tools together in one platform.
Some of our key accomplishments include:
- Building an AI-powered payment claim parsing and verification workflow.
- Successfully demonstrating a matching payment claim with a 95% confidence score and transaction-level evidence.
- Developing subscription analysis to identify recurring payments.
- Building Trap Scan to highlight price jumps and small-offer-then-big-charge patterns.
- Integrating the PayPal Developer Platform Sandbox into our application.
- Creating a unified dashboard with registration and login functionality.
Most importantly, we transformed an everyday financial problem into a working product demonstration.
What we learned
We learned that building an AI-powered financial application requires more than connecting an AI model to a user interface.
Reliable results depend on structured data, accurate transaction matching, clear evidence, and careful handling of uncertain patterns.
We also learned how to connect frontend and backend systems, work with databases, integrate external developer platforms, and design AI-assisted workflows that solve practical problems.
A key lesson was that unusual transaction patterns should be presented as warnings for further investigation, not automatically labelled as fraud.
What's next for Trace
We see Trace as a platform that can grow into a more complete financial clarity tool for freelancers and small businesses.
Our future plans include:
- Adding a complete, user-friendly bank statement upload experience.
- Expanding support for different banks and statement formats.
- Improving transaction matching and AI-assisted explanations.
- Enhancing subscription tracking and recurring expense insights.
- Improving PayPal invoice workflows and payment-status synchronization.
- Strengthening data privacy, account isolation, and security.
- Providing clearer evidence reports to help users investigate payment disputes.
Our long-term vision is to make financial transactions easier to understand, verify, and manage — all in one place.
Built With
- express.js
- google-gemini
- groq
- jest
- mongodb-atlas
- node.js
- paypal-sandbox-(invoicing-api
- react
- tailwind-css
- vercel
- vite
- webhooks)