Think Before You Grid-Search: Floor-First Triage for LLM Serving

Yihua Liu

LLM serving optimization typically benchmarks many configurations and reaches for heavy profilers when latency targets are missed. We argue for the reverse discipline: estimation i…