Zhangwei L.

Key Skills

Software

Google Cloud Vertex AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Task Types

Classification

Object Detection

AI Trainer Overview

AI Engineer | AI Relevancy Scoring & Audit Automation - HOMEFIELD. Brings 11+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Google Cloud Vertex AI. Education includes B.S., Tsinghua University (2012). AI-training focus includes data types such as Image and labeling workflows including Classification and Object Detection.

Labeling Experience

AI Engineer | AI Relevancy Scoring & Audit Automation - HOMEFIELD

Google Cloud Vertex AIImageClassificationObject Detection

Built a configurable Claude-based scoring workflow for URL evaluation rules including topical relevance, authority, spam risk, and link likelihood. Designed a framework normalizing six raw signals into a 0–100 composite score with traceable audit trails. Replaced manual SEO audits with an agent-driven process for prompt refresh, cache clearing, headless audits, and CSV export through FastAPI endpoints. Implemented Claude contact extraction with regex fallback, returning up to three ranked outreach names, titles, and emails per approved URL. Combined model outputs, rule-based checks, and structured responses for consistent scoring. Stack: Python, Claude, FastAPI, Pydantic, REST APIs, agent orchestration, structured outputs, CSV pipelines.

2022 - 2024

AI Developer | Autonomous Patient Support Agent - Bizzarro Agency

Google Cloud Vertex AIImageClassification

Built a Claude-powered healthcare support agent using RAG over product documentation, ticket triage, intent classification, and secure account and billing tool calls. Designed escalation rules for unclear, sensitive, or high-risk cases. Delivered the support experience on mobile with an Expo application backed by LangChain, Pinecone, and FastAPI. Reduced first-response time by 78%, lowered median resolution time by 54%, and increased CSAT by 22 points in one quarter. Balanced automation with safe human handoff for regulated healthcare workflows. Stack: Python, Claude, LangChain, FastAPI, Pinecone, RAG, Expo, React Native, PostgreSQL, REST APIs.

2020 - 2023

Education

Work History