Skip to main content

Welcome to Seshat

Welcome to Seshat, the Web3 primitive for predictive features. As the frontier of decentralized technology advances, the need for reliable, efficient, and accessible predictive analytics grows. Seshat is designed to meet this need by providing a comprehensive suite of tools tailored specifically for Web3 applications.


At Seshat, our mission is to transform how predictive features are developed and deployed within decentralized applications (dApps). We tackle the critical bottlenecks encountered by developers in the Web3 ecosystem, providing innovative solutions that enhance efficiency and reduce latency in predictive analytics.

Key Challenges in Web3 Predictive Feature Development

Developing predictive features for Web3 involves several unique challenges:

  1. Training Dataset Preparation: Efficiently preparing training datasets from on-chain data is crucial yet complex due to the variability and decentralized nature of the data.

  2. Lack of Ready-to-Use Models: The Web3 environment lacks a robust library of off-the-shelf models and pipelines tailored for on-chain data, making it difficult for developers to implement predictive features quickly.

  3. Deployment and Maintenance: While numerous technologies can train new predictive features, there remains a significant gap in streamlined deployment and ongoing maintenance of these features.

Seshat's Dual-Engine Solution

Seshat Overview

To address these challenges, Seshat introduces a dual-engine approach:

Training Dataset Engine (TDE)

  • Streamlined Data Collection: Automated processes to collect and preprocess on-chain data efficiently.
  • Automated Transformation: Advanced algorithms to transform raw data into a structured format ready for analysis.
  • Enhanced Security and Privacy: Ensuring data integrity and security throughout the preprocessing stage.

Real-Time Predictive Engine (RPE)

  • Off-the-Shelf Predictive Features: Access to a library of pre-built models specifically designed for Web3 applications.
  • Real-Time Inference: Capabilities to deliver predictions with minimal latency, crucial for real-time applications.
  • Robust Scalability: Designed to handle increasing data volumes without compromising performance.

Benefits of Seshat's Technology

  • Reduced Development Time: From weeks to hours, significantly speeding up the go-to-market for new features.
  • Cost Efficiency: Reduces the overhead associated with developing custom predictive models from scratch.
  • Scalability and Flexibility: Adapts to various blockchain environments and scales with the growth of your dApp.