Deployment & MLOps
Get your model off the laptop and into production.
Serving, scaling, monitoring, A/B testing, cost control. Cloud reference architectures on AWS, GCP, and Hugging Face — with the boring-but-critical bits like rollbacks and shadow traffic.
12h
Duration
10
Lessons
1.7k
Learners
Path map
Lessons unlock as you complete the previous one. Your progress is saved on this device.
Lesson 1
From notebook to production — the gap
9m35 XPLesson 2
Inference servers — vLLM, TGI, Triton, SGLang
12m50 XPLesson 3
Cloud GPUs — picking the right machine
10m40 XPLesson 4
Containers & immutable deployments
10m40 XPLesson 5
Autoscaling & traffic patterns
10m40 XPLesson 6
Cost optimization that actually moves the needle
11m45 XPLesson 7
Monitoring — what to actually watch
10m40 XPLesson 8
Shadow traffic, canaries, and A/B tests
11m45 XPLesson 9
CI/CD for ML pipelines
9m35 XPLesson 10
Capstone: design a production stack
14m70 XP