Bridging Diagnosis and Repair. Zero Guesswork.
Fixie AI
Timeline
Dec 2025 - Feb 2026
TEAM
Individual
MY ROLE
Behavioral Analysis
UX Strategy
UI & Interaction
Rapid Prototyping
AI Workflow
This case study focuses on closing the information gap between technicians and users through AI diagnostics. Discover how Fixie AI optimizes repair efficiency, boosting FTFR and reducing lead times.
GYUHO Lee © 2026
Interact
Users don't just want a fast fix. They want a transparent process.
01. Highlights
Process Support Guide:
Maximized FTFR,
Minimized Lead Time.
AI Diagnosis
From Guesswork to Data
Previously, users faced long wait times, relying on contacting technicians.
Fixie AI turns vague symptoms into clear, actionable diagnostics the moment a problem occurs.
DYNAMIC MATCHING
From Waiting to Visibility
Making the repair process visible with real-time updates and transparent timelines.
FAST RESPONSE
From Urgency to Action
Helping technicians review urgent jobs, prepare estimates, and respond with less friction.
02. Process
Human-Led,
AI-Accelerated Workflow.
Challenge
Logic Gaps & Generic Output
AI-generated plans often default to average solutions. They follow familiar patterns instead of adapting to nuanced contexts. In complex, human-centered scenarios, AI reasoning alone shows clear limitations.
Solution
AI for speed. Human for judgment.
I structured a workflow where human judgment guides key decisions, while AI accelerates execution. Clear intervention points maintain control and keep the process both systematic and adaptable. This is not about replacing human thinking, but knowing when to intervene.
Macro Framework
Phase 1.
Brain (Human)
Role - Empathy, Definition
Output - Problem Statement
Phase 2.
Spark (AI)
Role - Ideation, Variations
Output - Raw Layouts
Phase 3.
Hand (Human)
Role - Logic Check, Edge Cases
Output - Final Polish
4 Step AI–Human Collaboration
A structured execution model where AI accelerates production and humans safeguard strategic reasoning.
Step 1.
Research & Synthesis
Human: Conduct contextual interviews and gather qualitative signals to define initial direction.
AI: Analyze large-scale unstructured data to surface recurring pain points.
Outcome: Reduced bias and improved early-stage validation.
Step 2.
Information Architecture
Human: Define information priorities based on business goals and task flow.
AI: Detect logic gaps and edge cases across user scenarios.
Outcome: A resilient UX backbone with minimized blind spots.
Step 3.
Rapid Visual Exploration
AI: Generate fast low-fidelity UI explorations.
Human: Strategically reframe structures based on contextual realities and constraints.
Outcome: Production time reduced, focus shifted to meaningful adjustments.
Step 4.
Prototyping & Refinement
Human: Refined interactions based on psychological intent (clarity, urgency, trust signals).
AI: Provided feedback on interaction patterns and timing for smooth flows.
Outcome: A high-fidelity prototype that reduces cognitive friction.
03. Problem Space
Status: Pending. No ETA.
In traditional repair systems, users are often left in a state of “Pending.”
The real issue is not the breakdown itself, but the uncertainty built into the repair process.
While working at a restaurant in Berlin, I saw how opaque workflows and communication gaps turned even minor repairs into prolonged uncertainty.
UX Breakdown
Key Friction Points
Behavioral Pattern Analysis
Core Insight
Key Friction Points
Unclear Communication
Once an issue was reported, users often lacked clarity on the next step, response timing, or how their problem was being understood and handled.
Uncertain Timelines
Users often had no reliable sense of when a technician would arrive or when the issue would be resolved.
Lack of Real-Time Feedback
After the initial report, progress remained largely invisible to users.
How the Process Breaks Trust
When structural gaps turn into user impact.
💬
Language & Trust
Technical jargon and opaque procedures distance users from the process, weakening confidence in service quality.
“I wasn’t sure what was happening, but I had to pretend I understood.”
⏳
Process Anxiety
Not knowing when or how the issue will be resolved creates ongoing stress beyond the initial breakdown.
“I keep checking my phone. It feels like the whole day is on hold.”
📉
Operational Fallout
Lack of timely information and repeated technician visits disrupt daily operations and revenue.
“If one small part is missing, I lose another day of business.”
An information blind spot
"Users know the steps, but have no idea what is actually happening."
Quick Insight
"It’s not a lack of will.
It’s the burden of manual tasks and missing data."
👨🏻🔧 Technician Interview
The Anatomy of Friction
This wasn’t just a psychological issue — it was a visibility failure in the system. Mapping the journey revealed two critical uncertainty peaks where the system failed to provide clarity.
Peak 1. The Confirmation Gap (Scheduling)
No clear confirmation at each stage. Users are forced to track progress themselves.
Peak 2. The Predictive Failure (On-site)
No real-time inventory sync. Technicians arrive without required parts. Repeat visits extend downtime and operational loss.
What Users felt
“When are they actually coming?”
“How long is this going to take?”
“Will it be fixed properly?”
“Do I need to keep following up?”
The Core Insight
From 'Fragmented Steps' to a 'Synchronized Loop'
When users turn to AI or others for answers, it is not because they are impatient. It is because they feel uncertain.
The issue is not slow service. It is the lack of reassurance.
Strategic Definition
The solution was redefined beyond a simple matching system into a predictable communication platform that removes the black box between diagnosis and repair.
Design Principle
'Zero Guesswork'
Users should never have to guess what is happening.
I reduced visual noise and made key information (status / time / steps) immediately clear. The goal was not smarter AI, but clearer information that enables confident action.
Integrated Strategy Map
Bridging Friction & Solutions
1:1 matrix unifying business impact and design principles to prevent narrative gaps.
The Logic
Behind Solution
1.
Operational Inefficiency
Targeting a "Zero-Guesswork" Experience
Challenge: Information Asymmetry
Technicians often arrive without enough information about the issue.
Strategy: AI Pre-Diagnosis [The Translator]
Converts user descriptions into clear, structured data so technicians are prepared before arrival.
Goal: Maximizing FTFR
Replaces blind visits with better-prepared service.
2.
Market Bottleneck
Improving Match Speed Through Incentives
Challenge: Supply-Demand Imbalance
Matching delays during peak hours or for urgent tasks, leading to lost opportunities.
Strategy: Dynamic Time-Deal [The Matchmaker]
Highlights high-value jobs in real time to encourage faster responses.
Goal: Reduce Delays in Matching
Ensures urgent needs are met more quickly.
3.
Process Uncertainty
Building Trust by Minimizing "Anxiety Costs"
Challenge: High Anxiety Costs
Users feel stressed due to a lack of visibility during the repair process.
Strategy: Real-time Status Tracker [The Zen Master]
Provides a clear, trackable timeline that shows progress and builds confidence.
Goal: Build a Trust Loop
Encourages repeat use by making the process transparent and predictable.
04. Strategy
AI Strategy & 3 Pillars
: From Uncertainty to Action
AI doesn't replace technicians; it replaces uncertainty.
Why AI?
Users already rely on AI to interpret problems in everyday situations. I introduced AI at the critical gap between Diagnosis and Action, where user anxiety peaks. Here, AI does not replace expertise.
It structures incomplete signals so users and technicians can operate with a shared understanding.
The AI Bridge : Structuring the Process
Step 1.
The Problem (Uncertainty)
User: High anxiety driven by "Blind Guesswork" about the problem's cause.
Technician: Inefficient, "Unprepared Visits" due to a systemic lack of site data.
Step 2.
AI Intervention (Structuring Signals)
Data Structuring: Converting fragmented, natural language inputs into Actionable Data.
Pre-Diagnosis Report: Providing experts with real-time guides to ensure a "First-Visit Fix."
Status Synchronization: Real-time syncing of the technician’s timeline for Full Process Visibility.
Step 3.
Human Action (Confident Execution)
User: Gaining "Controlled Reassurance" through transparent, real-time updates.
Technician: Arriving as a "Prepared Expert" with the exact parts and tools needed.
3 UX Design Pillars
Three strategic pillars to bridge the information gap through AI-driven design.
01.
The Translator
AI Data Structuring
Turning messy user input into clear, actionable signals.
Replacing blind visits with data-driven preparation to improve first-time fix rates.
Prepared Visits, Not Guesswork
02.
The Matchmaker
Dynamic Incentives
Adjusting incentives in real time to balance supply and demand.
Encouraging faster responses for urgent tasks during peak hours.
Reducing Market Congestion
03.
The Zen Master
Timeline Clarity
A clear, trackable timeline that builds trust and reduces uncertainty.
Reducing uncertainty to naturally drive repeat usage and long-term loyalty.
Building a Trust Loop
05. Exploration
Merging AI Logic with
Human Craft
Design Goal
To accelerate early design exploration with Google Stitch and enable greater focus on structural thinking and problem-solving.
AI Exploration,
Human Judgment
Google Stitch
Generated diverse IA-based layouts using AI to expand the exploration space. Compared and validated multiple structures to identify the most intuitive information architecture.
My Role
Redefined visual priorities and information hierarchy based on service logic. Refined AI-generated drafts to improve data clarity and real-world accessibility.
🍌 The Translator
From Repair Confusion
to Clear Action
Problem
Users face error codes, but are expected to diagnose problems themselves. This leads to confusion, misdiagnosis, and unnecessary visits.
AI Strategy
Instead of interpreting error codes, users describe issues in natural language.
AI translates this into structured diagnostic signals.
AI Assistant
Fixie
Analyze and diagnose the issue with specialized technical AI
Connect users with professional technicians when needed
Before matching users with a technician, the system proposes self-repair options.
This prioritizes user autonomy and potential cost savings.
How does the technician see this?
The diagnostic report is designed around these principles.
Confident Action
The report translates complex AI data into a clear, prioritized task flow, enabling immediate and confident action.
Zero Blind Visits
Pre-identified error codes and required parts increase first-time fix rates, eliminating unnecessary return visits.
Task Classification with codes
The AI-generated UI initially organized tasks by customer name, but this approach would become difficult to manage as task volume increases.
Unlike B2C apps that prioritize names or locations, technician dashboard required a more systematic indexing structure to manage multiple tasks without confusion.
I structured tasks using unique job codes (e.g., Job #4092), enabling technicians to quickly identify and manage multiple cases.
🍋🟩 The Matchmaker
Supporting Technician
Workflows
Independent technicians often rely on manual workflows and unstable customer acquisition.
To address this, I introduced a ride-hailing–inspired model, integrating automated matching and structured workflows directly into the dashboard.
Two modes, Standard and Urgent, structure workflow around Urgency.
Decision order: Urgency → Value → Fit
Urgency Drives Earnings
“ASAP” turns urgency into priority, keeping the marketplace fluid.
Made financial incentives visible upfront, encouraging faster acceptance through reduced platform fees.
Technicians: Immediate visibility of higher earnings
Customers: Faster matching and service
Deep Dive: Resolving UX
Conflicts
In usability testing of the main task management tab, the AI-generated UI revealed critical interaction conflicts that static views failed to surface.
Interaction Conflicts
Map vs. Card Gestures
Overlapping scroll areas made one-handed interactions unreliable.
Reduced Map Visibility
Large cards reduced map visibility, making routes harder to scan.
The interface needed to adapt to real field behavior, not just visual layout.
View AI Draft → Final UI
Separated Interaction Zones
Horizontal scrolling separated task browsing from map interaction, reducing gesture conflicts.
Adaptive Viewport
Once navigation starts, the task card minimizes to keep the route visible while remaining easy to revisit.
🍋 The Zen Master
Operational Transparency
The problem wasn’t only the repair itself, but the invisible process around it. A shared timeline made each step visible, reducing uncertainty for both users and technicians.
Design Decisions
Shared Visibility
Aligned user-facing progress with technician actions, creating a single source of truth and reducing uncertainty.
Reduced Workflow
Compressed technician updates into four core actions, enabling fast, reliable input in real-world conditions.
PING System
Our shop owner pointed out a recurring issue: customers struggled to receive timely updates without repeated phone calls.
"Can technician status be checked in just a few taps?"
Side Effects :
More options ➔ More hesitation ➔ More confusion
Working with the shop owner, I removed unnecessary options and simplified the flow to the three most essential questions.
Response Windows & Accountability
When a technician does not respond within 10 minutes, the system steps in automatically.
Response Incentives
Responding within 10 minutes preserves the urgency fee benefit. Delays automatically forfeit it.
Reliability Signals
Repeated delays weaken reliability signals, reducing visibility in future matching.
Digital Handshake
To close the repair process with clarity and trust, the system introduces a digital confirmation flow between technician and customer.
In Germany, invoices are often sent several days after the work is completed, without prior notice, leaving customers surprised and uncertain about the final cost.
As a result, the service often lacks a clear and smooth sense of closure.
Interview findings showed that technicians could provide a rough estimate after identifying the issue. This created an opportunity to align expectations early in the process.
So I designed the system to allow users to understand an early cost estimate before the repair began.
Once the repair is complete, the technician finalizes the work and submits:
AI-summarized repair report
Final invoice
By structuring the billing flow around early alignment and final confirmation, the system makes the closing stage easier to understand and accept.
Just a Few Extras
Personalized Service Logic
A clear entry point that guides users into the right flow based on their role.
Everything, Organized
Seamlessly managing every repair step, from active tracking to historical records.
See full breakdown
A Clear Final Step
Before paying, customers can review the final cost, completed work, and technician proof in one clear flow.
06. Design Impact
What the System
Now Enables
As this project is still in the MVP stage, live metrics are not yet available. Instead, prototype-based user testing was used to evaluate where the redesigned workflow could reduce delay, uncertainty, and friction. While the sample size was small, repeated observations revealed three meaningful patterns.
Faster Diagnostic Alignment
After being matched with a technician, the time to receive an estimate was reduced from 11–14 days to under 30 minutes.
Less Back-and-Forth, Faster Diagnosis
By replacing phone and email ping-pong with a structured AI intake flow, unnecessary back-and-forth was reduced and diagnostic feedback became significantly faster, shortening the process from 5–7 days to near real-time.
Closing the Process, Faster and Clearer
Replacing Germany’s traditional invoice-based payment flow with in-app payment reduced service closure time from an average of 18 days to under 5 minutes.
AI Synergy & Efficiency
Integrating AI as a logical partner accelerated key stages of the design workflow, enabling faster exploration and deeper logic validation.
Early mockup production 3-4 weeks ➔ 1.5 weeks
Systemic Scalability
Beyond individual repairs, designing a resilient ecosystem for professional service logic.
B2B Logic Expansion
Evolving into a SaaS platform for repair agencies to manage larger workforces and more complex service operations.
Data-Driven Ecosystem (Predictive Maintenance)
Using accumulated repair logs to anticipate recurring issues and enable more proactive service planning.
Vertical Domain Growth
Extending specialized diagnostic logic into high-stakes domains such as HVAC and Electrical Engineering.
What Mattered Most
Reflections
Field-Driven Logic: The UI was shaped around real-world technician constraints such as thumb-zone reach, movement, and one-handed use in the field.
AI–Human Synergy: AI accelerated structure and exploration, while human judgment refined clarity, trust, and decision-making.
Restoring Market Trust: Fragmented repair touchpoints were redesigned into a clearer service journey, making the final step easier to understand and trust.
Domain over Aesthetics: Meaningful solutions began with deeply understanding local service realities, not just making interfaces look polished.
System Thinking & Prompting: In the AI era, stronger product logic and clearer prompting became essential design skills.
Critical Discernment: The designer’s role was not only to use AI, but to identify weak logic, challenge assumptions, and refine outputs into reliable decisions.
One Last Thought.
The real value wasn’t AI alone.
It was how AI and human judgment worked together
to turn uncertainty into structure.
"NEXT ARCHIVE"
nooding
DIRUNI
