Yilin Zhao
Sunnyvale
Summary
Senior Principal Machine Learning Engineer / Al Engineer with 8+ years of experience building production Al systems for cybersecurity and loT. Expertise in LLM-powered platforms, agentic Al systems, and hybrid retrieval architectures (vector + graph + enterprise data). Inventor of 10+ patents in LLM systems, IoT intelligence, and Al-driven security analytics.
Overview
9
9
years of professional experience
Work History
Senior Principal Machine Learning Engineer / AI Engineer
Palo Alto Networks
01.2024 - Current
- Architected domain-aware LLM platform (RAG + knowledge graph + enterprise data) enabling reasoning over code, documentation, and telemetry.
- Built production LLM agent system for IoT debugging and root-cause analysis, integrating MongoDB, Redshift, and internal services via LangGraph (ReAct).
- Designed multi-agent workflows (LangGraph, MCP, ADK) with tool-based reasoning and API-first data access.
- Developed hybrid retrieval system (dense + BM25+ HYDE + vector DB + graph) improving accuracy for complex technical queries.
- Built LLM-powered network design agent (NGFW) with automated topology generation (draw.io) and multi-step planning workflows.
- Led IoT analytics platform (Luminous) for device identification, vulnerability analysis, and knowledge systems (Devicepedia).
- Designed LLM-based vulnerability prioritization system using EPSS, device impact, and remediation scoring.
- IoT Security
Principal / Manager / Senior Staff MLE
Palo Alto Networks
09.2019 - 01.2024
Machine Learning Engineer/Data Scientist
ZingBox
01.2017 - 09.2019
- Developed ML systems for IoT device discovery and behavioral anomaly detection using large-scale network telemetry.
- Acquired by Palo Alto Networks
Education
M.S. - Statistics
University of Illinois at Urbana-Champaign
Urbana-ChampaignB.S. - Mathematics
Dalian University of Technology
Technical Expertise
RAG, Tool-Augmented LLMs, LangGraph, LangChain, Google ADK, MCP, A2A, Multi-Agent Systems, Hybrid Search (Vector + BM25+ Graph), Knowledge Graphs (Neo4j/Cypher), Qdrant, Python, SQL, MongoDB, PostgreSQL/Aurora, Kubernetes, AWS, Redshift, Pub/Sub
Patents
10+ patents across LLM systems, agent architectures, IoT intelligence and AI driven security analytics.
Representative:
- LLM-based MCP Communication Anomaly Detection
- Adaptive Agent & Function Access Control for Agentic Systems
- IoT Knowledge-Based Chatbot System
- AI-Driven Vulnerability Exploitation Assessment
- Automated PCAP Analysis & Information Extraction
Timeline
Senior Principal Machine Learning Engineer / AI Engineer
Palo Alto Networks
01.2024 - Current
Principal / Manager / Senior Staff MLE
Palo Alto Networks
09.2019 - 01.2024
Machine Learning Engineer/Data Scientist
ZingBox
01.2017 - 09.2019
M.S. - Statistics
University of Illinois at Urbana-Champaign
B.S. - Mathematics
Dalian University of Technology