Space and Time (SxT) is a verifiable compute layer that enables trustless data processing for Web3 applications. SxT combines a decentralized data warehouse, a novel zero-knowledge proof system, and an AI-powered SQL engine to provide scalable, secure, and easy-to-use data solutions for smart contracts, LLMs, and enterprises.

SxT consists of three main components: the data warehouse, the Proof of SQL, and the AI SQL.

The data warehouse is a hybrid transactional/analytic (HTAP) system that provides both low-latency in-memory cache and high-performance GPU-accelerated analytics. The data warehouse supports SQL as the query language and allows users to join, transform, and analyze data from various sources, including:

• Indexed blockchain data from major chains such as Ethereum, Polygon, Sui, Avalanche, and Sei

• Offchain data loaded from databases, data warehouses, object storage, applications, or game servers

• Realtime data streams from sensors, IoT devices, oracles, or other Web3 services

The data warehouse is decentralized and distributed across a network of nodes that store, process, and verify data. Users can access the data warehouse through the SxT client, which provides a user-friendly interface for data ingestion, query execution, dashboard creation, and API publishing.

Proof of SQL is the novel zero-knowledge proof system that guarantees the correctness and integrity of the data and queries in the data warehouse. Proof of SQL leverages zk-SNARKs, a cryptographic technique that allows one party to prove to another that a statement is true without revealing any information beyond the validity of the statement.

Proof of SQL works as follows:

• When a user submits a query to the data warehouse, the query is executed by a subset of nodes called executors, who generate the query result and a corresponding proof.

• The proof is then sent to another subset of nodes called verifiers, who check the validity of the proof without having to re-execute the query or access the underlying data.

• If the proof is valid, the verifiers sign the proof and broadcast it to the rest of the network, creating a consensus on the query result and the proof.

• The user can then receive the query result and the proof, and use them for various purposes, such as:

  • Sending tamperproof query results to smart contracts in a trustless way

  • Publishing query results directly onchain using SxT's native token, STX

  • Sharing query results with other users or applications with cryptographic guarantees

AI SQL is the AI-powered SQL engine that enhances the user experience and productivity of the data warehouse. AI SQL provides several features, such as:

• SQL Prompt: a natural language interface that allows users to write SQL queries using plain English

• SQL Autocomplete: a code completion tool that suggests SQL keywords, table names, column names, and operators as users type

• SQL Formatter: a code beautifier that formats SQL queries according to best practices and standards

• SQL Optimizer: a query optimizer that rewrites SQL queries to improve their performance and efficiency

• SQL Visualizer: a data visualization tool that generates charts and dashboards from SQL queries with a single click

AI SQL leverages deep learning models and natural language processing techniques to understand the user's intent, generate SQL code, and provide feedback and suggestions. AI SQL aims to make SQL more accessible, intuitive, and enjoyable for users of all skill levels.