Inspiration
FinBud was inspired by the chaos of today’s digital world, where buying stuff is way too easy and overconsumption is basically the default. With constant ads, endless trends, and one‑click checkouts, we’re all spending more than we mean to. We built FinBud to help individuals slow down, question the urge, and make choices that feel good, and not just good in the moment. Now you can turn mindless spending into intentional choices, helping you keep your money, your sanity, and your closet from overflowing.
What it does
FinBud helps you pause before you make an impulse purchase. Enter your wage, select a mode (friend, therapist, financial advisor), and paste the link of your intended purchase. Depending on the mode, FinBud will give you a calm reflection, real financial advice, or a chaotic reality check.
How we built it
We built FinBud by starting with Python on the backend, where we calculated the verdict for whether a product should be bought. We defined decision rules with the help of AI to make the verdict smarter and more consistent. Along the way, we experimented with OCR, but due to time constraints and complexity, we pivoted to web scraping to keep the project moving. That combination of flexibility, problem‑solving, and AI‑assisted logic is what brought FinBud to life. We also integrated an LLM using Mistral to explain the logical reasoning behind each verdict.
Challenges we ran into
One of our biggest challenges was implementing OCR. We quickly realized it was more complex and time‑consuming than we could handle within our timeline. Instead of getting stuck, we pivoted to web scraping, which allowed us to keep the core functionality while still making the user experience smoother.
Accomplishments that we're proud of
We’re proud to have built a real, functioning program. What started as an overly ambitious idea with a huge UI vision turned into something simple, intentional, and effective. FinBud may not be flashy, but it carries a message we care about: slowing down, spending smarter, and pushing back against the overconsumption culture we’re all stuck in.
What we learned
We learned very quickly that time constraints matter. Getting stuck on one problem for too long can stall the entire project, so we had to pivot fast and stay flexible. Instead of obsessing over a single issue, we focused on finding workable solutions and keeping our momentum. That shift in mindset is what allowed us to finish our app—progress over perfection, every time.
What's next for FinBud
We’re just getting started. Next, we want to add features that make FinBud even more intuitive and helpful—like letting users enter the quantity of items they’re considering and giving the interface a smoother, more modern feel. We aim to further explore OCR technology so users can simply screenshot their checkout page or shopping cart instead of copying and pasting links. Looking ahead, we plan to expand the session history and build AI that learns spending patterns over time, offering smarter, more personalized advice. FinBud will keep evolving into a tool that not only helps you pause, but truly understands how you spend.
Built With
- beautiful-soup
- mistralai
- python
- requests
- scraperapi
- streamlit
Log in or sign up for Devpost to join the conversation.