Getting Started
Learn how to discover, understand, use, and work with resources from Axomiya AI Foundry.
Axomiya AI Foundry provides open Assamese language resources, datasets, and infrastructure for developers, researchers, educators, and the wider community. This guide explains where to begin and how to work with the resources responsibly.
Start with the Datasets
Explore the Datasets section to understand what resources are currently available.
Each dataset page provides information about the purpose of the dataset, what it contains, how it can be used, and relevant limitations.
Before using a dataset in a project, read its documentation and check the applicable licensing and usage information.
Understand the Data
Do not treat a dataset as an unquestionable source of truth. Dataset quality depends on how the underlying data was collected, prepared, reviewed, and maintained.
Review available metadata, descriptions, provenance information, and known limitations before using data for research or production systems.
Using Foundry Resources
Developers can use Foundry resources to experiment with Assamese language technology, build applications, conduct research, and create new language resources.
Researchers should document which datasets and versions were used in their work so that results can be reproduced as accurately as possible.
If you publish work using Foundry resources, follow the citation guidance provided in the documentation.
Before You Build
Check the dataset documentation and usage terms.
Understand the intended purpose and limitations of the resource.
Consider privacy, consent, bias, safety, and potential misuse where applicable.
Keep track of the dataset version used in your project.
Explore datasets
Browse available Assamese language datasets and understand their purpose before using them.
Read data governance
Understand the principles we use to encourage responsible data development and use.
Learn how to cite
Credit the datasets and resources you use in research, publications, and other public work.
Need Help?
If something is unclear, check the relevant dataset documentation first. If you identify missing information, an error, or an opportunity to improve the documentation, consider contributing a correction or improvement.