11 section
Infrastructure & MLOps
The operational layer: where self-hosting starts beating API pricing, how you autoscale GPU capacity, and how a prompt change gets through CI without shipping a quality regression.
What is in here
Two chapters that assume the model already works and ask how you run it. Start with LLM Infrastructure for the deploy, serve and scale decisions, then CI/CD for the release process that keeps a prompt edit from becoming an incident. Both are written for whoever carries the pager, not for the prototype.
01
8 min
LLM Infrastructure
API versus self-hosted at the volume where the economics flip, plus serving architecture, GPU autoscaling, cost controls, alerting and a disaster-recovery plan.
infrastructuredeploymentcost
02
6 min
CI/CD for LLM Applications
Shipping prompt and model changes safely: staged eval suites, quality gates that block a regression, canary rollout, and a rollback that actually works.
deploymentevaluationinfrastructure