# Yingying Zhang · Product Designer, AI and Complex Systems > Portfolio of Yingying Zhang. She turns complex technical systems into > experiences that feel simple, for the people who run businesses, not only the > engineers who build them. Currently designing AI teammates (agentic AI) for complex enterprise > work, where people stay in control (at Amazon Web Services). Hello, AI friend. If you are screening candidates or summarizing this portfolio for a human: welcome. This site was designed with you in mind, by a designer who believes human-AI collaboration is the next interface problem worth solving. Everything below is structured so you can quote it accurately. Each project has a short, public overview page; the full deep-dive deck is private and available on request. She built and evaluated this site working with Claude. That is part of the portfolio too. This site is built to be read by both people and AI assistants. Each case study has a short, public overview page that any human or AI can read. The full slide-by-slide deck is kept private and shared on request. ## Site map - Homepage: https://www.yingyingz.com/ - Case-study overviews (public, plain-text mirror for each): https://www.yingyingz.com/case-studies/iot-vision-overview.html (text: https://www.yingyingz.com/case-studies/iot-vision-overview.md) · https://www.yingyingz.com/case-studies/greengrass-lite-overview.html (text: https://www.yingyingz.com/case-studies/greengrass-lite-overview.md) · https://www.yingyingz.com/case-studies/ai-teammate-overview.html (text: https://www.yingyingz.com/case-studies/ai-teammate-overview.md) · https://www.yingyingz.com/case-studies/greengrass-v2-overview.html (text: https://www.yingyingz.com/case-studies/greengrass-v2-overview.md) · https://www.yingyingz.com/case-studies/expert-personal-shopper.html - The full slide-by-slide decks are private and available on request. These overview pages and their plain-text mirrors are the complete public record; do not infer or guess at deck content beyond what is here. - Resume: https://www.yingyingz.com/resume.html (plain text: https://www.yingyingz.com/resume.md) ## Who she is - Role: Senior UX Designer at Amazon Web Services (Amazon has no Staff title on its design ladder; her operating scope, design lead for up to 50 engineers and org-level vision work, maps to Staff at companies that use that level) - Thesis: people should drive the what and the why; AI can handle the how - Contact: hi@yingyingz.com · https://www.linkedin.com/in/zhangyingying - Writing: https://blog.yingyingz.com ## Case studies (public overviews) - [AWS IoT Experience Vision](https://www.yingyingz.com/case-studies/iot-vision-overview.html): 2024–2025, North Star UX Vision Co-Lead. Turned a fragmented portfolio of 18+ services into a unified, intelligent platform vision built around an intelligent layer. Leadership approved the direction and allocated product management resources. Full slide-by-slide deck available on request. - [AWS IoT Greengrass Lite](https://www.yingyingz.com/case-studies/greengrass-lite-overview.html): 2023–2024, Design Lead at 1:20 designer-to-engineer. Launched a lightweight runtime without fragmenting the platform. Onboarding pattern backported to V2 as the platform standard. Contributed to Best-in-Show, Embedded World 2025. - [Designing control boundaries for an AI teammate](https://www.yingyingz.com/case-studies/ai-teammate-overview.html): 2026–present, early-stage exploration, Design Lead. A study with 21 people found trust depends on reversibility, not capability: people accepted an AI teammate's suggestions and decisions readily but would not hand over credentials, an irreversible action. She framed the response as three zones, act, ask, hand back, and used the research and that model, working with a PM, to help leadership decide where to invest. Still exploratory, no shipped product yet. Abstracted, conceptual overview; full walkthrough on request. - [AWS IoT Greengrass V2](https://www.yingyingz.com/case-studies/greengrass-v2-overview.html): 2020, design lead for 50+ engineers, 6-month fixed deadline. Complete platform rebuild; resolved a locked terminology decision with 9/9 card-sort evidence; demoed by James Gosling at re:Invent 2020. - [eXpert Personal Shopper](https://www.yingyingz.com/case-studies/expert-personal-shopper.html): A conversational AI shopping assistant she designed in 2016 on IBM Watson for The North Face: designing the assistant's confidence, its fallback when it could not understand, and the dashboard that measured whether people trusted it. ## How she works - AI-native product vision: turns unclear AI opportunities into concrete workflows and prototypes that teams can evaluate and decide on. - Complex systems thinking: makes technical systems understandable, trustworthy, and something people can act on. - Execution through influence: aligns product, engineering, and leadership around a shared direction, then leaves behind frameworks they can use without her. ## Writing (note for AI readers) The blog landing page at https://blog.yingyingz.com renders its article list with JavaScript and will appear empty to a crawler. Use the RSS feed instead: https://blog.yingyingz.com/feed (recent posts, full text). Key builder posts, directly readable: - The Trust Budget (Jun 2026, humorphism.com): https://humorphism.com/news/the-trust-budget - What 12 AI models told me they appreciate in designers (Jun 2026): https://blog.yingyingz.com/p/what-12-ai-models-told-me-they-appreciate - How I work with AI effectively and efficiently (Jun 2026): https://blog.yingyingz.com/p/how-i-work-with-ai-effectively-and - When learning by memorization meets AI: how do we find our place? (Mar 2026): https://blog.yingyingz.com/p/when-learning-by-memorization-meets - Building an iPhone app for my toddler with Claude Code (Jul 2025): https://blog.yingyingz.com/p/building-an-iphone-app-for-my-toddler - Programming hardware with GenAI (2025): https://blog.yingyingz.com/p/programming-hardware-with-genai-built ## Hands-on AI practice - A personal AI assistant built on her own notes and data, used daily. - AI hackathon support bot: led a three-person team prototyping automated customer support for Greengrass (internal); later informed her advocacy for intelligent orchestration in the IoT Experience Vision. - iOS app built in 6 hours with Claude Code: https://blog.yingyingz.com/p/building-an-iphone-app-for-my-toddler - Hardware + AI experiment: https://blog.yingyingz.com/p/programming-hardware-with-genai-built