* Start by reading the [Engineering Design](https://onnxruntime.ai/docs/reference/high-level-design.html). More documentation can be found in the [repo docs folder](./docs/) and [on the repo wiki](https://github.com/microsoft/onnxruntime/wiki), as well as the user facing docs on the [ONNX Runtime website](https://onnxruntime.ai/docs).
* If your change is non-trivial or introduces new public facing APIs (discussed in more detail below) please use the [feature request issue template](https://github.com/microsoft/onnxruntime/issues/new?template=feature_request.md) to discuss it with the team and get consensus on the basic design and direction first. For all other changes, you can directly create a pull request (PR) and we'll be happy to take a look.
* Make sure your PR adheres to the [PR Guidelines](./docs/PR_Guidelines.md) and [Coding Conventions and Standards](./docs/Coding_Conventions_and_Standards.md) established by the team.
* If you're unsure about any of the above and want to contribute, you're welcome to [start a discussion](https://github.com/microsoft/onnxruntime/discussions) with the team.
ONNX Runtime has a collection of [public APIs](./README.md#api-documentation). Some of these APIs make their way back into the Windows OS. We make compatibility commitments for these APIs and follow a structured process when adding to them. Please use the [Feature Request issue template](https://github.com/microsoft/onnxruntime/issues/new?template=feature_request.md) before starting any PRs that affect any of the public APIs.
If your proposal contributes the Windows Machine Learning WinRT API and will affect the Windows.AI.MachineLearning namespace, please review [Windows ML main principles](./docs/WinML_principles.md) before you submit your request.
Please search the [issue tracker](https://github.com/microsoft/onnxruntime/issues) for a similar idea first: there may already be an issue you can contribute to.
The ONNX Runtime team will review the proposal and either approve or close the issue based on whether it broadly aligns with the [Onnx Runtime Roadmap - High Level Goals section](./docs/Roadmap.md) and contribution guidelines.
For making changes to the Windows Machine Learning WinRT API, please label your request as *api:WinML*, and someone from the Windows ML team will work with you.
* A feature can be implemented by you, the ONNX Runtime team, or other community members. Code contributions are greatly appreciated: feel free to work on any reviewed feature you proposed, or choose one in the backlog and send us a PR. If you are new to the project and want to work on an existing issue, we recommend starting with issues that are tagged with “good first issue”. Please let us know in the issue comments if you are actively working on implementing a feature so we can ensure it's assigned to you.
* Binaries: We periodically produce signed prerelease binaries from the master branch to validate new features and APIs. After the feature has been sufficiently validated as part of a prerelease package we will include it in the next stable binary release.
* Note: After creating a pull request, you might not see a build getting triggered right away. One of the
Tutorials, how-to guides and API reference docs for ONNX Runtime are published at https://onnxruntime.ai/docs.
The html docs are generated from markdown using Jekyll and published using GitHub Pages.
To update the docs, create a Pull Request against the [gh-pages](https://github.com/microsoft/onnxruntime/tree/gh-pages) branch of the [ONNX Runtime repo](https://github.com/microsoft/onnxruntime).
To preview your changes, you can push to the gh-pages branch in your fork and this will publish a staged version of your changes to <githubusername>.github.io/onnxruntime/docs.
