Inspiration

As a retail investor and a FinTech student, I noticed that most market participants react emotionally to financial news headlines. News such as central bank decisions, inflation data, or global geopolitical events often create panic or hype, even though many retail investors do not fully understand the actual market impact behind these headlines.

I was inspired to build a tool that does not predict stock prices or give investment advice, but instead explains the cause-and-effect relationship between news and market behavior in simple language. The goal was to help retail investors make more informed and rational decisions rather than reacting impulsively.

What the Project Does

MarketSense AI is a Gemini-powered application that analyzes financial and macroeconomic news and explains its potential impact on equity markets.

Given a news headline or article, the application:

Classifies the news (macroeconomic, sector-specific, or company-specific)
Identifies affected market sectors
Explains short-term (1–7 days) and long-term (6+ months) implications
Provides India-specific market context (NIFTY, Sensex, and major sectors)
Clearly explains the reasoning behind each impact using beginner-friendly language

The project focuses purely on **education and understanding, not investment recommendations.

How I Built It

The application was built using Google AI Studio and the Google Gemini API.

I designed a structured prompt that guides Gemini to:

Perform semantic understanding of financial news
Reason through cause-and-effect market dynamics
Generate clear, structured explanations suitable for retail investors
Avoid financial advice, price predictions, or trading signals

Google AI Studio was used to deploy the application as a live, interactive web experience, allowing users to paste news and instantly receive structured market impact insights.

Challenges I Faced

One of the main challenges was ensuring that the AI output remained explanatory rather than advisory. Financial AI tools often drift into predictions or recommendations, which I intentionally avoided to stay focused on education and responsible usage.

Another challenge was designing prompts that balanced financial accuracy with simplicity, so that the output remained understandable even for beginners with limited market knowledge.

What I Learned

Through this project, I learned:

How to use Gemini’s reasoning capabilities for real-world financial analysis
The importance of prompt engineering in shaping AI behavior
How to translate complex macroeconomic concepts into clear, user-friendly explanations
How AI can be responsibly applied in fintech without replacing human judgment

This project reinforced my interest in building responsible, explainable AI tools for financial education.