www.infoq.com
Presentation: Scaling LLM-Based Ranking Systems for Latency-Critical Search & Recommendation Workloads
Sundara Raman Ramachandran shares how LinkedIn productionizes LLM ranking under strict P99 latency SLAs. He discusses distilling a 7B model down to 400M parameters, compressing job contexts into single-token embeddings for a 10x throughput leap, and eliminating hidden system bottlenecks - from custom flash attention masks to Python GIL workarounds, gc.freeze
Original reporting from www.infoq.com.
Read original at www.infoq.com