Zheng Zhang

“Built and shipped a real-time AI phone agent for small businesses that handles bookings/FAQs/messages using streaming ASR, an LLM with tool-calling, and TTS; deployed to production for multiple paying customers. Demonstrates strong applied LLM reliability practices (tool-first grounding, retrieval, hard-negative testing, and production monitoring) and experience orchestrating multi-step AI workflows with Airflow, Prefect, and AWS Step Functions.”

“Built a production LLM-driven archival assistant that turns large, low-quality scanned handwritten files (120+ pages) into structured datasets, overcoming context-window and hierarchy challenges with a two-phase LLM + rules pipeline and reaching 98.1% accuracy (Gemini-2.5 Flash). Also orchestrated a large human-in-the-loop effort with 78 archivists, producing 2,400 high-quality annotations in 4 days via detailed rubrics and support.”