Router that virtually distributes inference across connected devices in the home.
Go
docs: add roadmap and development team to README
Seth Schneider394049464a0a
diff --git a/README.md b/README.mdindex 0f0a724..9638244 100644--- a/README.md+++ b/README.md@@ -252,24 +252,59 @@ Component references, for when you already know what you are looking for:See the [releases page](https://github.com/NVIDIA/Personal-AI-Router/releases)for what changed in each release.-## Where PAIR is going--We have plenty of ideas about where to take PAIR, and no fixed commitments about-which of them land or when. If you have a thought about the product's direction,-something that would make it more useful to you, a workflow it does not support-yet, or a use we have not considered — we would like to hear it. Open an issue-and start the conversation.--**Routing is the clearest example.** Today PAIR ships a single scheduling policy-that combines queued work with a coarse, smoothed GPU-utilization signal. It does-not consider GPU model, available memory, model warmness, or how expensive a-request looks, which still makes it a better fit for similar machines than a-highly mixed cluster. Making that smarter, and likely letting you choose a-policy, is something we want to do — and hearing which of those signals matters-on your hardware is exactly the kind of input that would shape it.--Feedback from people running PAIR on their own hardware is more useful to us than-any plan written in advance.+## Roadmap++These are features we want to add to PAIR. This list is a direction for the+project, not a commitment to delivery or implementation order. Community+feedback and contributions will help shape priorities.++### Platform support++- [ ] Full support for Windows ARM64 systems.+- [ ] DGX Station support.++### Engines and integrations++- [ ] llama.cpp support.+- [ ] vLLM support.+- [ ] EXO support.+- [ ] ComfyUI integration.+- [ ] Unsloth support.+- [ ] Tailscale integration.++### Routing and clusters++- [ ] Route OpenAI-compatible API requests across different inference engines.+- [ ] Inference request queuing.+- [ ] Increase number of scheduler variables for better QoS.+- [ ] A "cluster as a node" view.+- [ ] KV cache-aware scheduling.++### Usability and reliability++- [ ] Requester-only clients that send work without hosting an inference engine.+- [ ] Launch PAIR automatically on system startup.+- [ ] Configure engine launch options and environment variables from PAIR.+- [ ] Stability and reliability improvements informed by real-world use.++Have a feature request or a workflow you want PAIR to support? Open an+[issue](https://github.com/NVIDIA/Personal-AI-Router/issues) and tell us how you+would use it.++## Development Team++NVIDIA team members working on PAIR:++| Name | GitHub | Role |+| --- | --- | --- |+| Noah Tervalon (Terve) | [@Noah-Tervalon-Nvidia](https://github.com/Noah-Tervalon-Nvidia) | PAIR Developer - Community Lead |+| Chris Kelsey | [@ckelseynv](https://github.com/ckelseynv) | PAIR Developer - UI/UX Lead |+| Sherief Farouk | [@sherief-nv](https://github.com/sherief-nv) | PAIR Developer - Scheduling, Team Lead |+| Preston Goode | [@nv-pgoode](https://github.com/nv-pgoode) | PAIR Developer - Engine Management Lead |+| Kaylee Lubick | | PAIR Developer - Security Lead |+| Lucas Brodzinski | [@LB-NV](https://github.com/lb-nv) | PAIR Technical Program Manager |+| Ambrish Dantrey | | PAIR Engineering Manager |+| Seth Schneider | [@NV-sschneider](https://github.com/NV-sschneider) | PAIR Product Manager |## Contributing and governancediff --git a/docs/architecture.mdx b/docs/architecture.mdxindex f7e7ffe..5e3bca6 100644--- a/docs/architecture.mdx+++ b/docs/architecture.mdx@@ -451,7 +451,7 @@ value and stays in the normal ranking.Improving this, and likely offering a choice of policies, is planned work. Whichof these gaps matters most depends on the hardware people actually run, soreports from real deployments are more useful than guesses. Refer to-[Where PAIR is going](../README.md#where-pair-is-going).+[Roadmap](../README.md#roadmap).## Engines and Portsdiff --git a/docs/known-issues.mdx b/docs/known-issues.mdxindex 3ed3a0d..6853f0c 100644--- a/docs/known-issues.mdx+++ b/docs/known-issues.mdx@@ -43,7 +43,7 @@ The full list of gaps is in[Scheduler Limitations](architecture.mdx#scheduler-limitations).Improving this is on the roadmap, and knowing which signals matter on yourhardware would help. Refer to-[Where PAIR is going](../README.md#where-pair-is-going).+[Roadmap](../README.md#roadmap).## PAIR Does Not Detect a Stuck Service Automatically