What’s your problem statement?

Investments - There's a million sources of financial advice and investment recommendations on the internet. Can people have their own personalized AI-powered one?

The problem investmentGURU solves

This investment platform helps users manage their finances with AI-driven insights and personalized recommendations, making investment decisions easier, smarter, and more secure. People can use this platform to keep track of all their investments in one place, making it simpler to see their financial status at a glance. By collecting information on a user’s current investments, financial goals, risk tolerance, and preferences, the platform creates a profile that reflects each individual’s unique financial needs and ambitions. The platform’s algorithms then analyze this profile alongside the user's existing investments, finding trends and identifying areas that are performing well or may need adjustment. This way, users receive valuable insights into the strengths and weaknesses of their portfolios without needing to be financial experts.

One of the most helpful features is the platform’s personalized recommendations. Based on the user's profile and current market conditions, it provides suggestions that align with their goals and risk levels. For example, someone with a low-risk tolerance might receive recommendations focused on safer investment options, while those with higher risk tolerance could be advised on investments that might offer higher returns. This customization not only saves time for users by presenting relevant options but also increases the potential for making profitable choices that are aligned with their financial goals.

Challenges we ran into

One specific hurdle in building this project was running LLaMA directly on our PC, which wasn’t feasible due to its high computational requirements. My hardware hit its limits with such intensive AI processes. To overcome this, we set up a virtual machine (VM) to handle LLaMA’s demands. The VM provided the necessary resources for LLaMA to run smoothly, and using it allowed us to continue developing without hardware constraints.