Home
Menu
Home
Start here
AI
Kubernetes
Networking
About
|
LIGHT
DARK
Optimization
2026
Pruning LLM models: cutting without amputating
Jun 5, 2026
Prefix cache: hit rate engineering to go from 15% to 75%
Jun 5, 2026
Optimising prefill in vLLM: the knobs your TTFT will not forgive
Jun 5, 2026
Optimising decode in vLLM: squeezing every token out of small hardware
Jun 5, 2026
Knowledge Distillation: teaching a small model to think like a big one
Jun 5, 2026
Instrumenting vLLM with OTel: measuring what the optimisations really do
Jun 5, 2026
Batch sizing in vLLM: the two-hour grid search worth weeks of hardware
Jun 5, 2026
Semantic cache in RAG: the receptionist with a photographic memory
Jun 4, 2026