Spice AI

What is Spice AI?

Spice AI is an enterprise-ready platform that provides the data and AI infrastructure needed to build applications using time-series data. It lets developers combine real-time and historical time-series data, custom data processing (ETL), machine learning model training, and predictions within a single connected AI backend service.

Standout features include high-quality indexing of web3 data from ecosystems such as Bitcoin, Ethereum, and Uniswap, machine learning pipelines, a model registry for sharing trained models, and the ability to run your own code on every piece of data. The goal is to make the core tools for building intelligent applications accessible to developers, even without deep data science or machine learning expertise.

Who created Spice AI?

Spice.ai was founded by Luke Kim and Phillip LeBlanc. Luke Kim was a founding manager and co-creator of Azure Incubations at Microsoft. Phillip LeBlanc has experience building large distributed systems from his time at GitHub and Microsoft. The company officially launched on May 26, 2022, with the vision of making it easier and faster to create AI-driven applications across fields like health and security.

What is Spice AI used for?

• Building data and time-series AI applications that rely on data over time.
• Composing real-time and historical time-series data with custom ETL.
• Machine learning training and inferencing, with streamlined pipelines.
• Simplifying blockchain interactions — no need to work with JSON RPC APIs or smart-contract ABIs.
• Eliminating complex infrastructure for obtaining web3 data or running blockchain nodes.
• High-quality web3 data indexing with block-level data for Bitcoin, Ethereum, Uniswap, and more.
• Querying blockchain data with SQL to get answers in seconds.

Who is Spice AI for?

• Data scientists looking for efficient data handling.
• Machine learning engineers needing robust pipelines and model management.
• Blockchain developers who want to integrate data easily.
• Web3 developers building next-generation applications.

How to use Spice AI?

Getting started is straightforward:

1. Import the SDK into your Node.js, Go, Python, or Rust project with just three lines of code, giving you access to petabyte-scale data platforms for extracting features, training models, and making predictions.
2. Use machine learning pipelines pre-connected to high-quality time-series data. Share trained models with colleagues or explore community models through the model registry.
3. Query at enterprise-grade speed using Apache Arrow APIs, which support rapid querying and fetching of millions of records.