← curriculum/phase 17 · Infrastructure And Production

edit on github ↗

← all lessons in phase

phase 17 · 28 lessons

01Managed Llm Platforms
02Inference Platform Economics
03Gpu Autoscaling Kubernetes
04Vllm Serving Internals
05Eagle3 Speculative Decoding
06Sglang Radixattention
07Tensorrt Llm Blackwell
08Inference Metrics Goodput
09Production Quantization
10Cold Start Mitigation
11Multi Region Kv Locality
12Edge Inference
13Llm Observability
14Prompt Semantic Caching
15Batch Apis
16Model Routing
17Disaggregated Prefill Decode
18Vllm Production Stack Lmcache
19Ai Gateways
20Shadow Canary Progressive
21Ab Testing Llm Features
22Load Testing Llm Apis
23Sre For Ai
24Chaos Engineering Llm
25Security Secrets Audit
26Compliance Frameworks
27Finops Llms
28Self Hosted Serving Selection

‹

›

phase 17 · lesson 03 of 28

Gpu Autoscaling Kubernetes

03-gpu-autoscaling-kubernetes·full lesson folder ↗

Inference Platform Economics

complete & next →

Vllm Serving Internals