Trace an LLM Request Through a Paged-KV Inference Engine
Company: AMD
Role: Software Engineer
Category: ML System Design
Difficulty: medium
Interview Round: Onsite
Overview: Trace LLM inference from request admission through prefill and decoding, explaining paged KV block tables, continuous batching, cache lifetime, and latency tradeoffs.
Read the full AMD Software Engineer interview experience this question came from