Zengxiao (Henry) He

AI/ML Engineer · Stanford M.S. EE · Agentic Systems


About

I’m an AI/ML engineer and builder focused on turning language models into reliable systems that can plan, use tools, remember context, and execute real operational workflows.

My work spans agent architecture, retrieval and memory systems, full-stack AI products, and automation platforms. Recently, I’ve built AI agents for supply-chain operations, customer engagement, dispatch optimization, and long-context memory.

I like working close to messy, high-friction problems where better software can change how people actually operate. Outside of work, I’m usually playing fingerstyle guitar, on a tennis court, or learning performance driving.


Experience

Co-Founder & Full-Stack Engineer

Glowia AI

Nov 2025 — Mar 2026

Built a conversational AI agent for MedSpa customer engagement, automating Instagram DM intake, appointment booking, deposits, and operator monitoring through a React dashboard and configurable LLM pipelines.

Software Engineer Intern — AI/ML

Oracle

Jun 2025 — Sep 2025

Developed and deployed an AI-powered automation agent for Oracle's supply chain platform, integrating Node.js, Puppeteer, Slack API, OCI Queues, Docker CI/CD, and Llama-4 root-cause generation to cut backport processing time by 40%.

Research Assistant

Stanford University — Prof. Tom Lee

Dec 2024 — Apr 2026

Designed an associative memory system for LLM agents, replacing standard RAG with graph-based spreading activation, multi-stage retrieval, keyword extraction, and time-decay scoring for long-context agent recall.

Teaching Assistant (CS 295: Software Engineering)

Stanford University

Dec 2024 — Apr 2026

Teaching and mentoring students on building robust software systems alongside Prof. Sara Achour. Designing assignment workflows, reviewing LLM-generated code for correctness, and providing project feedback to students.

Full-Stack Engineer Intern

Sumixer

Apr 2024 — Sep 2024

Built an AI-powered dispatch optimization platform using FastAPI, React, PostgreSQL, geospatial scoring, LLM-assisted ranking, and a real-time operator dashboard for local home-service operations.


Projects

Associative Memory for LLM Agents

A graph-based agent memory system that combines semantic similarity, spreading activation, YAKE keyword extraction, and time-decay scoring to improve multi-hop recall.

AI Lead-to-Booking Agent

A conversational agent that qualifies inbound leads, answers service questions, schedules appointments, collects deposits, and gives operators real-time visibility into conversion performance.


Skills

PythonTypeScriptJavaReactNext.jsFastAPINode.jsLLM agentsRAGFunction callingPrompt pipelinesDockerPostgreSQL

Writing

All writing →

Education

Stanford University

M.S. in Electrical Engineering, AI & Machine Learning

Sep 2024 — Jun 2026

GPA: 3.8/4.0. Teaching Assistant for CS 295 Software Engineering.

Central South University

B.Eng. in Software Engineering

Sep 2020 — Jun 2024

GPA: 3.9/4.0. Core Member, National Innovation & Entrepreneurship Base.