Building high-quality AI-enabled edtech requires an adaptive understanding of open licensing, structured pipelines that involve humans, and leveraging the open-source benchmarks, datasets, and models already available to the community. If you are looking to learn more about building more responsible and effective AI-enabled edtech for education, here are resources from The Learning Agency to explore:
Where to Find Open-Licensed Texts for Open Educational AI Projects The Core Idea: If you are building an AI tutor, text simplifier, or reading assessment tool, you need massive amounts of clean text – this post guides you in finding available texts.
Choosing The Right Annotation PlatformThe Core Idea: Before your data can work with an AI model, a human has to clean, structure, and label it.
Designing a Rigorous Data Annotation PipelineThe Core Idea: Bad data in, bad AI out. This post moves past the software and focuses heavily on human operations and the importance of quality control.
Free Datasets & New Educational Datasets For AI Research
The Core Idea: These final two posts feature deep dives into massive, fully documented, open-source datasets that are ready to drop directly into your research or product models.