Investara

Smart Planning. Wise Spending. Confident Investing

Comment

Inspiration

The idea for Investara came from my personal experience during my internship and a growing curiosity about how finances are managed. I realized that understanding investment risks, tracking portfolio performance, and making smart financial decisions shouldn't be limited to experienced investors or large institutions. As someone transitioning from computer science to fintech, I found most wealth management tools overwhelming and not beginner-friendly. This led me to build something simple and approachable — a tool that turns complex financial data into clear, actionable insights. Almost like having your own Personal Financial Assistant (PA) that helps you plan smart and spend wisely.

What it does

Investara currently features two smart agents that guide users through the early stages of understanding the stock market:

Visualization Agent

  • Uses a fixed list of Nifty 50 stock tickers.
  • Fetches and updates real-time stock data from Yahoo Finance.
  • Displays graphs to visualize individual stock performance over time. Identifies:
    1. Top Gainers – stocks with the highest positive return.
    2. Top Losers – stocks with the biggest drop in price.
    3. Bullish, Bearish or Doji (neutral) candles based on daily price movement.
    4. Helps users quickly assess market sentiment and stock volatility.

Finance Guide Agent -Answers beginner-friendly questions about investing and the stock market -Covers common topics like:

  1. “How do I get started with stocks?”
  2. “What is the difference between mutual funds and equities?”
  3. “What should I know before investing?” -Together, these agents help users learn and interact with market data in an intuitive and supportive way.

How we built it

We built Investara using the following stack and tools:

  • Frontend: HTML, CSS, JavaScript
  • Backend: Python with FastAPI for high-performance APIs
  • Data Handling: pandas and NumPy for processing CSV/JSON data
  • Visualization: Plotly and Matplotlib for dynamic and static charting
  • Deployment: Hosted on Render.com, with version control through GitHub The overall architecture is modular, with separate layers for data ingestion, processing, visualization, and user interaction — making it scalable, clean, and easy to maintain.

Challenges we ran into

  • Time Pressure: Building two agents in a short timeframe meant we had to prioritize core functionality and delay advanced features.
  • Data Accuracy: Making sure real-time stock data was reliable and handled gracefully when APIs failed or returned incomplete results.
  • Debugging Deployment: Faced unexpected issues related to environment setup, dependency mismatches, and platform compatibility.
  • Feature Creep: It was tempting to keep adding ideas, but we had to stay focused on building a stable, working model

Accomplishments that we're proud of

  • Successfully created two functional and interactive agents
  • Enabled live data visualization with customizable Nifty stock tickers
  • Built a beginner-friendly finance Q&A assistant using AI
  • Designed a structure that's flexible and ready for expansion
  • Delivered a solution that is both technical and user-centric

What we learned

  • Financial data is messy and requires careful handling of edge cases, especially across timeframes
  • Using FastAPI and modular patterns helped speed up development and debugging
  • Good visualization isn’t just about pretty charts — it's about meaningful context and filters
  • Product thinking matters: knowing what not to build is as important as the features you do include

What's next for Investara

  • Add OAuth login and secure user accounts
  • Integrate with real brokerage platforms like Zerodha or Upstox
  • Implement AI-generated alerts and recommendations based on portfolio performance
  • Add goal-based financial planning (retirement, vacation, emergency fund)
  • Redesign the interface to be responsive and mobile-first

Built With

Share this project:

Updates