Investing isn’t just about numbers — it's about future. Many beginner investors make impulsive decisions driven by emotions like excitement, fear, or stress. We wanted to build something that not only teaches sound trading strategies but also helps users understand how their emotional state affects their decisions. Inspired by the need for emotional intelligence in finance, SAVI (Smart Assistant for Value Investing) was born.

SAVI is a beginner-friendly web-based stock trading simulator that allows users to buy or sell listed stocks based on simple market data and predictions. It analyzes historical trends to suggest whether an investment decision is likely to be profitable. What sets SAVI apart is its integration with a hardware haptic device that monitors the user's physiological signals (like pulse or grip pressure) to infer emotional and excitement levels. Before a trade is confirmed, SAVI evaluates whether the user is in the right emotional state to make sound financial decisions — teaching them to invest not just with logic, but with self-awareness.

One of the most challenging aspects was integrating the Gemini API with our machine learning model, as it required seamless communication and compatibility between both components.

Finding a suitable dataset and defining a consistent happy flow for the application took significant time and experimentation due to data limitations and workflow inconsistencies.

With one team member from Computer Science and the other from Computer Engineering, we had to merge our distinct skill sets to tackle both software development and hardware integration effectively.

We struggled to measure haptic responses using the sensors available but adapted by analyzing body cues as an alternative solution.

Given the 24-hour timeframe of the hackathon, it was pretty challenging to bring these ideas together.

Built an end-to-end application that blends data, emotion, and education into a single interactive platform.

Successfully integrated hardware feedback into a real-time decision-making system.

Created a tool that not only simulates investing but fosters emotional intelligence.

The intersection of tech and behavioural finance is powerful — teaching users how they invest can be just as important as what they invest in.

Emotional state plays a significant role in decision-making, and even simple biometric cues can offer valuable insight.

Building user-centric tools that teach through experience is a great way to promote long-term learning.

Integration of more advanced machine learning models for stock prediction and emotion detection.

Expanding to mobile platforms for broader accessibility.

Increasing the difficulty levels in the Gamified learning version through badges, challenges, and community investment competitions.

Integration of Chatbot so that user has a better understanding if stuck anywhere without the wait time.

We can take this further by also developing the hardware components to involve more haptic gestures and signals to enable people with disabilities.