The problem Pulse solves

Crypto trading is challenging not only because of financial risk, but also due to the overwhelming amount of information that must be analyzed continuously. The steep learning curve discourages newcomers, while even experienced traders often lack the time to perform deep market research. This creates a gap between opportunity and execution.

Our solution leverages advanced LLM models to bridge that gap by processing large volumes of data, identifying meaningful patterns, and transforming them into actionable insights. By reducing cognitive load and automating complex analysis, we make trading decisions faster, safer, and more accessible for a wide range of users.

As real-world use cases, Echo monitors Bitcoin in real time and predicts its potential movements, giving users a practical tool to react quickly and capitalize on opportunities. Pulse, on the other hand, scans Ethereum-based tokens to detect potential rug pulls and other risks, providing early warnings that improve safety and protect investments. Together, these tools demonstrate how AI-driven analysis can simplify crypto trading while enabling users to trade smarter, safer, and with more confidence.

Challenges we ran into

Building both Echo and Pulse came with unique challenges. For Echo, one of the hardest parts was figuring out which data points from Bitcoin should be collected periodically to make meaningful predictions about its next moves. Beyond just gathering the data, training an LLM model to understand and analyze these signals required careful design and experimentation.

For Pulse, the complexity was even higher. Continuously scanning Ethereum blocks to detect newly deployed tokens in real time demanded both efficiency and accuracy. After collecting these tokens, we had to identify relevant risk factors for each one and evaluate whether they applied, before training our LLM to perform reliable rug-pull analysis. Managing this pipeline while keeping it scalable and responsive was a major challenge we had to overcome.

Smart Contract Address(es)

0x982a13b0E95EFf8F5776cD1E3BB959135DD4A0c4