Explosion develops developer tools such as spaCy and Prodigy for natural language processing, machine learning, and data annotation, enabling efficient text analysis and model training. Their solutions address the challenges of data labeling and model deployment, facilitating the creation of robust AI applications across various industries.
Funding
$9M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Developing natural language processing (NLP) applications requires specialized tools for efficient text analysis, data annotation, and model training. Existing solutions often lack the flexibility to integrate with the latest large language models (LLMs) or the modularity needed for production deployment.
Solution
Explosion provides a suite of open-source developer tools, including spaCy and Prodigy, designed to streamline the creation of NLP applications. spaCy is an industrial-strength NLP library that offers pre-trained models and a consistent API for various NLP tasks, while Prodigy is a scriptable annotation tool that accelerates data development and model training. The company's tools facilitate the integration of LLMs into NLP pipelines, enabling developers to prototype quickly and distill knowledge into smaller, task-specific components for efficient and data-private deployment. By emphasizing modularity and human-in-the-loop distillation, Explosion helps teams translate business problems into machine learning solutions, while maintaining modularity, transparency, and data privacy.
Target Audience
The primary audience includes AI and machine learning developers, data scientists, and NLP engineers who need efficient, scalable, and customizable tools for building NLP applications.
Features
- spaCy: An open-source NLP library with pre-trained models, a consistent API, and support for custom components
- Prodigy: A scriptable annotation tool for data development and model training, featuring active learning and efficient annotation workflows
- LLM integration: Tools for incorporating LLMs into spaCy pipelines, enabling fast prototyping and knowledge distillation
- Modular architecture: Components designed for modularity, transparency, and data privacy
- Human-in-the-loop distillation: Techniques for distilling LLM knowledge into smaller, task-specific models
- Support for custom tasks: Ability to define custom tasks for entity and relationship extraction
- Pre-configured workflows: Streamlined workflows for common NLP tasks, such as NER, text classification, and relationship extraction
- Scalable pipelines: Tools for building scalable NLP pipelines that can process large volumes of text data