Welcome to the AI-for-Education GitHub page, a collaborative hub for developing and sharing AI tools and resources aimed at revolutionizing education in low and middle-income countries (LMICs).

We want you to use these tools to democratise access to AI in education, ensuring every student and teacher, regardless of their background, can benefit from these technologies.

• Pedagogy Benchmark: Code for the paper Benchmarking the Pedagogical Knowledge of Large Language Models. This benchmark uses questions testing pedagogical knowledge from teacher qualification exams. The dataset contains 1143 questions spanning a range of subjects (or categories), age groups (e.g. primary or secondary school level) and pedagogical subdomains.
• FabData-LLM: A library for interfacing with various LLM APIs and tools for managing chatbots.
• FabData-LLM-retrieval: A full end-to-end platform for building a Retrieval-Augmented Generation system from a catalogue of documents.
• Lesson Plan Parse MBSSE: A python package for parsing the raw text of Sierra Leone MBSSE lesson plan pdfs into structured json and subsequently cleaning the text so that it is suitable for human consumption. Also contains public links to the parsed and cleaned lesson plans.
• Digitizing the Global Proficiency Framework: This repository contains a digitized version of the Global Proficiency Framework for Reading, including the example assessment items, to support and enable development of computational tools based on this framework.
• Automating alignment to the Global Proficiency Framework: This repository provides interactive code and scripts which build on our digitization of the GPF. Code is included for a few experimental application.
• FabData-parsedoc: A library for document text extraction and content parsing / summarisation / interpretation.
• Redis Stack Server ACI: A tool for automated deployment of a TLS-enabled Redis Stack server as an Azure Container Instance. This could be used, for example, to host a vector database for Retrieval Augmented Generation (RAG).
• And more: see the full list in repositories and watch this space as we publish more - or contribute yourself!

We encourage contributions from educators, developers, and AI enthusiasts. Check out our individual project pages for more details on how to contribute.

We are continually developing papers, presentations, and publications that share our work. You can access these on our website, https://AI-for-education.org with select works linked below.

• [Coming Soon]

Find tutorials, documentation, and guides in our various project repositories and on https://AI-for-education.org to help you get started with AI in education.

For support, questions, or collaboration requests, please contact support@ai-for-education.org.

Follow us on LinkedIn and Twitter to stay updated with our latest developments and events.

Our projects are open-source and available under the MIT License unless stated otherwise in individual repositories.