-
-
Code base
-
Chat.1
-
Database
-
Chat.2
Inspiration
All of us who worked on this project live together in a shared flat. And honestly, we kept forgetting small payments—someone paid for milk, someone covered an auto fare, someone loaned a few hundred rupees… and after a week nobody remembered who owed whom. This constant confusion made money management messy. That’s when we thought: “Why not build something that remembers all our financial conversations better than we do?” And that idea became our FinTech Bot.
What it does
Our FinTech Bot is an AI-powered WhatsApp assistant that manages your personal finances through simple chat messages. You can tell it:
Your salary
Your daily expenses
Loans you gave or took
Repayments
Savings goals
And it stores, organizes, and recalls everything instantly—no matter how old the conversation is. It also gives advice based on your income and spending patterns, tracks loans, generates summaries, and understands natural language using NLP. Unlike regular AI chat models that forget data after 2–3 messages, our bot remembers as long as the conversation exists.
How we built it
We used:
Python + Flask for the backend and webhook handling
Twilio as the messaging service to connect with WhatsApp
HTML, CSS, JavaScript, and Tailwind CSS for the frontend dashboard
Encryption to ensure complete privacy so no user can access another user’s data
NLP models to understand natural language messages and extract financial details
The system continuously logs chats, processes transactions, and intelligently updates the user’s financial record.
Challenges we ran into
Setting up and stabilizing the webhook took much more effort than expected.
The backend logic—especially accurate extraction of financial data—required multiple revisions.
Our server crashed a few times early on, but we fixed that.
The UI is still under development, and syncing frontend + backend is ongoing.
One major challenge we’re still working on is processing audio messages. Understanding and converting voice notes into structured financial data is tricky.
Accomplishments that we're proud of
The bot can track loans, expenses, income, and savings seamlessly.
It gives smart financial advice based on user salary and spending habits.
It maintains a long-term memory of conversations—something normal chatbots can’t do.
We implemented secure encryption, making sure all user data stays private and isolated.
We built a working AI finance companion entirely from scratch with a small team.
What we learned
We learned how powerful teamwork can be—everyone contributed a different skill. We also learned:
End-to-end integration of Twilio + WhatsApp
How to build and manage webhooks
NLP techniques for extracting financial data
Backend optimization, debugging crashes
Importance of privacy and encryption in fintech products
Most importantly, we learned that real-life problems often lead to the best tech solutions.
What's next for FinTech Bot ~ “Never forget any payment again.”
We learned how powerful teamwork can be—everyone contributed a different skill. We also learned:
End-to-end integration of Twilio + WhatsApp
How to build and manage webhooks
NLP techniques for extracting financial data
Backend optimization, debugging crashes
Importance of privacy and encryption in fintech products
Most importantly, we learned that real-life problems often lead to the best tech solutions.
Log in or sign up for Devpost to join the conversation.