Agent-driven staging, deployment, benchmarking, and optimization of open-weight models on Amazon SageMaker AI using portable SKILL.md contracts.
sample-sagemaker-agentic-model-deployment is a coding-agent workflow that takes an open-weight model from a public checkpoint to a deployed, benchmarked, and optimized endpoint on Amazon SageMaker AI. It is for automating model staging, serving-configuration selection, deployment, benchmarking, and optimization without manually writing the surrounding API and polling code. Its skills and instruction define the portable agent workflow.
These files are aws-samples/sample-sagemaker-agentic-model-deployment's own configuration. They tell Kiro, Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.
AGENTS.md A 8 tok .kiro/skills/sagemaker-benchmark/SKILL.md A 79 tok .kiro/skills/sagemaker-deploy/SKILL.md A 67 tok .kiro/skills/sagemaker-optimize/SKILL.md A 90 tok