Inspiration

Investment platforms today focus on telling users what to invest in, but rarely explain why. As students and young investors, we noticed how intimidating and opaque financial decision-making can be—especially for beginners who are expected to trust complex systems without understanding them.

FinSight AI was inspired by the idea that financial confidence comes from clarity, not blind automation. We wanted to build a platform that prioritizes transparency, education, and trust alongside intelligent recommendations.


What it does

FinSight AI is an explainable, risk-aware investment guidance platform that helps users understand their investment decisions.

The platform:

  • Collects user profile details such as age, goals, and risk appetite
  • Generates a personalized portfolio recommendation
  • Clearly explains why a specific allocation was chosen
  • Allows users to simulate “what-if” scenarios by adjusting risk and time horizon
  • Highlights ethical considerations through transparent logic and a dedicated Trust & Ethics layer

Instead of acting as a black box, FinSight AI empowers users to learn while investing.


How we built it

We built FinSight AI as a full-stack, modular MVP:

Frontend:

Next.js with TypeScript and Tailwind CSS for a clean, modern, and responsive UI

Logic Layer:

A centralized, rule-based recommendation engine designed to be explainable and easily extendable to machine learning models

Visualization:

Interactive charts and sliders to help users explore portfolio allocations and risk metrics

Backend (Prototype):

FastAPI setup to support API-driven recommendations and future scalability

The system was designed with clarity and extensibility in mind rather than over-engineering.


Challenges we ran into

  • Designing investment recommendations that were both simple and realistic
  • Translating financial logic into human-readable explanations
  • Structuring a clear end-to-end user journey without overwhelming the user
  • Balancing technical depth with ethical responsibility in a FinTech context
  • Managing time while building a polished UI and stable logic under hackathon constraints

Each challenge required careful trade-offs between complexity, usability, and trust.


Accomplishments that we're proud of (so far)

  • Building a connected end-to-end MVP with a clear user flow
  • Implementing an explainable recommendation system, not just static outputs
  • Creating an interactive scenario simulator that encourages informed decision-making
  • Including a dedicated Trust & Ethics layer—often overlooked in hackathon projects
  • Delivering a clean, professional, and user-friendly experience

What we learned

  • Explainability is just as important as accuracy in financial systems
  • Good UX can significantly improve user confidence in complex domains
  • Ethical considerations should be built into the product, not added later
  • Clear communication often matters more than complex algorithms
  • Building for real users requires thinking beyond just “making it work”

What's next for FinSight AI

  • Gradually transitioning from rule-based logic to ML-assisted recommendations
  • Integrating real-time market data for dynamic portfolio updates
  • Adding user authentication and portfolio persistence
  • Expanding compliance, risk disclosures, and regulatory safeguards
  • Scaling the platform as an educational and advisory tool for beginner investors

FinSight AI aims to grow into a platform that not only guides investments but also builds long-term financial literacy.

Built With

Share this project:

Updates