using-ml-production

This Skill routes your production concern—deployment, optimization, MLOps, or observability—to the correct sub-skill. State what you need (e.g., 'deploy to production' or 'optimize inference'), and it directs you to the relevant reference sheet (deployment-strategies, quantization-for-inference, model-serving-patterns, or production-monitoring-and-alerting).

Production routing covers model deployment strategies, hardware optimization, quantization for inference, model serving patterns, and monitoring. It classifies your concern into Model Optimization, Serving Infrastructure, MLOps Tooling, or Observability and points you to the matching guide.

Use this Skill when you're moving models to production but unsure which aspect to address first. It reduces cognitive load by classifying your concern and directing you to the right sub-skill, avoiding the guesswork of choosing between deployment, optimization, MLOps, or monitoring.

Yes. This Skill routes MLOps and observability concerns to experiment-tracking-and-versioning and production-monitoring-and-alerting reference sheets, covering workflow establishment and production observability setup.

Yes. The Skill routes inference optimization concerns to quantization-for-inference and hardware-optimization-strategies references, guiding you on reducing model size and latency in production.

This Skill includes safeguards to avoid routing non-production questions. If your concern falls outside deployment, optimization, MLOps, or observability, it won't route to production sub-skills.