project screenshot 1
project screenshot 2
project screenshot 3
project screenshot 4
project screenshot 5


AI-powered trade setups with verifiable and provable inference on Starknet


Created At


Project Description

Entropy is a platform that offers users trade suggestions generated by AI models. These AI models can be provided by us, and general users can also upload their trained models, which will be visible to others. Users can then subscribe to the models and run inferences for a particular token to learn the predicted values and suggested trade positions in order to maximize their financial gains based on market conditions.

Upon running an inference, the user will also receive a cryptographic proof of the inference that they can independently verify on their machines. This feature is essential to instil confidence in users who are utilizing the models, ensuring that the subscription fee is being effectively utilized and that they are not falling victim to scams

How it's Made

This project uses LSTM (Long Short-Term Memory) networks for price prediction and trade generation, integrated into a platform that utilizes Starknet and Cairo for computational integrity and verification. The platform allows users to upload their AI models for others to use, ensuring transparency and security through verifiable proofs.

Technologies Used

Starknet and Cairo:

Starknet: Utilized for its ability to provide scalable and secure proofs of computational integrity. This is critical for ensuring the accuracy and reliability of AI inferences made by the models. Cairo: Chosen for its powerful and flexible programming capabilities that enable the creation of provable programs. It prevents cheating and malfunction by keeping the computational process honest and verifiable

LSTM for Price Prediction:

Purpose: The core of our self-developed model uses LSTM networks to predict prices and generate trades for scalping and swing positions. Implementation: LSTM networks are well-suited for time series forecasting due to their ability to remember long-term dependencies and handle sequential data efficiently.

background image mobile

Join the mailing list

Get the latest news and updates