End-to-end orchestrator that takes a NEW model from a bare HuggingFace id to a perf-lake-published, validated CROSS-ENGINE (vLLM + SGLang) champion on B200/GB300. Scaffolds a per-model harness + evidence bundle (run-id == experiment-id), then drives gate-driven phases: deploy baseline -> 4-layer profile (zymtrace L1 / DCGM L3 / ncu L4) -> tune CROSS-ENGINE via the variant A/B (vLLM AND SGLang arms) -> quantize -> validate -> spec-decode train+validate -> bench the multi-workload suite -> championselect (baseline vs top-X, the obvious production pick) -> publishtolake. Pauses at every red gate. Every number defaults to DRAFT (a champion VERDICT needs the multi-workload + accuracy gates + L3 byte-grounding). Triggers on "optimize a new model", "bring up a model end-to-end", "model bring-up pipeline", "find the best perf for ", "run the full optimization pipeline", or any combination of "optimize / bring-up / end-to-end / pipeline / champion" with "new model / inference / vllm / sglang".
Install with
Claude CodeCodexCursorGitHub CopilotGemini CLI
Supported by the Atlas install prompt. Skill-specific requirements vary.
Skill stats
Skills1CategoryScience & researchStars0
About this skill
End-to-end orchestrator that takes a NEW model from a bare HuggingFace id to a perf-lake-published, validated CROSS-ENGINE (vLLM + SGLang) champion on B200/GB300. Scaffolds a per-model harness + evidence bundle (run-id == experiment-id), then drives gate-driven phases: deploy baseline -> 4-layer profile (zymtrace L1 / DCGM L3 / ncu L4) -> tune CROSS-ENGINE via the variant A/B (vLLM AND SGLang arms) -> quantize -> validate -> spec-decode train+validate -> bench the multi-workload suite -> championselect (baseline vs top-X, the obvious production pick) -> publishtolake. Pauses at every red gate. Every number defaults to DRAFT (a champion VERDICT needs the multi-workload + accuracy gates + L3 byte-grounding). Triggers on "optimize a new model", "bring up a model end-to-end", "model bring-up pipeline", "find the best perf for ", "run the full optimization pipeline", or any combination of "optimize / bring-up / end-to-end / pipeline / champion" with "new model / inference / vllm / sglang".
Quick start
Copy the install prompt. Use the green button above this README.
Send it to your agent. Paste it into your coding agent’s chat and send.
Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.
End-to-end orchestrator that takes a NEW model from a bare HuggingFace id to a perf-lake-published, validated CROSS-ENGINE (vLLM + SGLang) champion on B200/GB300. Scaffolds a per-model harness + evidence bundle (run-id == experiment-id), then drives gate-driven phases: deploy baseline -> 4-layer profile (zymtrace L1 / DCGM L3 / ncu L4) -> tune CROSS-ENGINE via the variant A/B (vLLM AND SGLang arms) -> quantize -> validate -> spec-decode train+validate -> bench the multi-workload suite -> champion_select (baseline vs top-X, the obvious production pick) -> publish_to_lake. Pauses at every red gate. Every number defaults to DRAFT (a champion VERDICT needs the multi-workload + accuracy gates + L3 byte-grounding). Triggers on "optimize a new model", "bring up a model end-to-end", "model bring-up pipeline", "find the best perf for <model>", "run the full optimization pipeline", or any combination of "optimize / bring-up / end-to-end / pipeline / champion" with "new model / inference / vllm / sglang".
About this listing
Category
Science & research
Reported license
MIT
Listing language (estimated)
English
Review status
Not individually reviewed
Agent compatibility
Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLI