Add governance process page to deepspeed.ai#8138
Conversation
Signed-off-by: Zhipeng Wang <zwanga@wustl.edu>
Introduce a governance process document at /governance/ on deepspeed.ai, tailored for DeepSpeed's TSC-based structure. Covers project values, maintainer roles, committer proposal process, decision making, and contribution policies. Includes TSC committers and core maintainers list. Signed-off-by: Zhipeng Wang <zwanga@wustl.edu>
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 8d07af91b1
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| @@ -0,0 +1,163 @@ | |||
| --- | |||
There was a problem hiding this comment.
The root AGENTS.md requires every new file to include the SPDX/DeepSpeed Team license header, but this newly added governance page starts directly with Jekyll front matter and has no license notice. This violates the repository's new-file policy and can cause license-compliance checks or reviews to reject the change; please add the required notice in a Markdown/Jekyll-safe form.
Useful? React with 👍 / 👎.
| 1. **Top performance**: System performance is our top priority. We monitor overheads, optimize kernels, and publish benchmarks. We never leave performance on the table. | ||
| 2. **Ease of use**: DeepSpeed must be simple to install, configure, and operate. We provide clear documentation, fast startup, clean logs, helpful error messages, and monitoring guides. Many users fork our code or study it deeply, so we keep it readable and modular. | ||
| 3. **Wide coverage**: DeepSpeed supports frontier models and high-performance accelerators. We make it easy to add new models and hardware. DeepSpeed + PyTorch form a simple interface that avoids complexity. | ||
| 4. **Production ready**: DeepSpeed runs 24/7 in production. It must be easy to operate and monitor for health issues. |
There was a problem hiding this comment.
I don't think this is applicable to DeepSpeed (or a training engine) unlike VLLM (inference engine).
| ### Advisory Board | ||
|
|
||
| DeepSpeed project leads consult with an informal advisory board that is composed of model providers, hardware vendors, and ecosystem partners. | ||
|
|
There was a problem hiding this comment.
This is currently not applicable unlike VLLM
3 participants
Summary
/governance/on deepspeed.aiTest plan
bundle exec jekyll serveindocs//governance/link appears in the sidebar navigation