Liz Yu

Lead Product Designer @EvolutionIQ working on complex system problems, driving innovation with strategic leadership, and making an impact with empathy-centered design.

Previously @Evisort, @Mark43, @U.S. Digital Response

 
 

Case studies ☟☟

 
 
 

EvolutionIQ

Medical Summary: AI & Human-in-the-loop Review

At EvolutionIQ, we’re on a mission to get people back to work. Claims handlers, however, are tasked with the arduous work of going through pages of documents to figure out what’s happening with the claim medically and support claimants in getting paid while out of work from an injury or illness. By using AI for tedious initial assessments, claim handlers can make key claims decisions and focus on personalized care. The future of medical documentation and patient management relies on balancing AI efficiency with human oversight for optimized speed and accuracy.

AI | 0-1 product | behavior modification | stakeholders

Want to learn more? Let’s get in touch

 

Mark43

Combined Search Experience

I worked with product leadership and growth to create a vision for combined search across our entire suite of six products. This included aligning stakeholders, building and testing a vision concept, leading the designers on my team, and working with cross-functional teammates to break down this idea into iterative, actionable projects.

Strategy | PLATFORM design | Stakeholder management | DESIGN LEAD

BG-bright blue.jpg
 

Mark43

Mobile Product: Vision to Reality

I was tasked to figure out what my company’s mobile strategy would be and what the opportunities were for first responders responding in the field. I led user research and market opportunity analysis, pitched the vision to the org, and led the team to build a 0-1 iOS & Android native mobile product. I’ve since launched customers with a leading mobile solution for first responder workflows in the field.

GROWTH | Research | STRATEGY | Android & iOS

BG 16.jpg
 
 

DESIGN challenge

Slack Huddles: AI Summary & Recording

Often, conversations happen in Slack, and then a subset of participants move to a huddle to continue the conversation. However, some team members who weren’t in the huddle see the thread where it left off with no conclusion and they are stuck with incomplete information or potentially incorrect information. I challenged myself to solve this problem to ensure everyone is in the loop on how a conversation ended.

AI TECHNOLOGY | UX/UI | prototype poc

BG 6.jpg
 
 

EVISORT

LLM Orchestration &
AI Enhanced Instructions

Super users can create custom LLMs specific to their industry needs to work with their document data at scale. This feature allows customers to improve their model accuracy by fine-tuning their LLM, system prompt, and AI-generated instructions according to their unique industry needs.

AI & LLM | innovation | Stakeholder management