Summary
Francis Fan is an AI Application Engineer with a decade of experience building production-grade LLM-powered systems at Tencent and Huawei, and a proven track record delivering the first commercial telecom LLM deployment. He designs robust agent loops, multi-model routing, and RAG pipelines—optimising cost and context reuse (e.g., boosting KV-cache hit rates from 50% to 80% and cutting token costs ~40%)—and has driven measurable gains in accuracy and throughput across telecom and data-centre products. Currently building QClaw, a desktop AI agent that enables natural-language computer control via WeChat, he also introduced AI-assisted developer workflows and automated RAG evaluation with Neo4j-backed hybrid recall. Comfortable leading cross-functional teams and shipping end-to-end solutions, he pairs hands-on engineering (prompt engineering, LoRA fine-tuning, MLOps) with pragmatic product instincts and is open to senior AI/ML engineering or product roles in New Zealand.
10 years of coding experience
Bachelor's degree, Computer Science , Bachelor's degree, Computer Science at University of Minnesota