Deep Detective

Inspiration

Cryptocurrency has become a highly relevant form of financial investment for millions of people worldwide. However, it can be difficult to assess the risk associated with different cryptocurrencies. We created a "Deep Detective" which is an analysis tool to help users better evaluate crypto assets. Users can search for a cryptocurrency, compare it with others, and add notes to track their thoughts. This helps people build more informed analyses as they make investment decisions.

What it does

Our platform uses AI along with external APIs to search the web, gather relevant information, and summarize key insights about different cryptocurrencies. It analyzes trends, news, and available data to generate an easy-to-understand risk assessment score. This helps users quickly evaluate potential risks without needing to sort through overwhelming amounts of information themselves. Users can compare different cryptocurrencies side by side, track their findings, and share their analyses with others, making research more collaborative and accessible.

How we built it

We built our project using Cursor as our development environment and experimented with multiple AI models, including MiniMax and Gemini, to power our analysis and summarization features. For gathering real-time information, we integrated Tavily to perform web searches and pull relevant data. The frontend was developed as a Vite-based website, allowing us to create a fast, modern, and responsive user interface. Together, these tools enabled us to combine AI-driven insights with live web data to generate meaningful crypto risk assessments.

Challenges we ran into

One of the main challenges we faced was limited usage credits for some of the tools and AI models we initially relied on. We started by using MiniMax and really enjoyed the experience and performance, but credit constraints made it difficult to sustain for a fully working product. As a result, we had to adapt our AI workflow and switch models to ensure the system could run reliably within our limits. This required us to be flexible with our technical decisions while still maintaining the quality of our analysis features

Accomplishments that we're proud of

We’re proud of our ability to adapt quickly and still build a working product under time and resource constraints. Throughout the project, we explored and learned new tools such as MiniMax and Tavily, which made the development process both exciting and educational. Gaining hands-on experience with these technologies helped us grow as builders, and we’re excited to continue learning and improving our skills in future projects.