ENTERPRISE AI SUITE
ENTERPRISE AI SUITE
ENTERPRISE AI SUITE
Enterprise AI Suite to provide seamless and efficient AI Research and optimize usage of computing resources
Details
Details
Details
Role:
Role:
Lead Product Designer
Design System
Service:
Service:
Enterprise AI Suite
AI Dev Tools
Date:
Date:
Jan. 2025 - Aug. 2025
Overview
Overview
Overview
Yennefer Suite is 5 in 1 enterprise AI platform that serves AI search to get valuable insights from massive data and manage an organization's AI projects. I was the only designer in this startup company and led the entire design thinking process from research to launch. Currently, Yennefer is being used by 9 enterprise organizations including Samsung Display, KT Telecom, Hyundai, Seoul National University, and the State University of New York.
Yennefer Suite is 5 in 1 enterprise AI platform that serves AI search to get valuable insights from massive data and manage an organization's AI projects. I was the only designer in this startup company and led the entire design thinking process from research to launch. Currently, Yennefer is being used by 9 enterprise organizations including Samsung Display, KT Telecom, Hyundai, Seoul National University, and the State University of New York.
Yennefer Suite is 5 in 1 enterprise AI platform that serves AI search to get valuable insights from massive data and manage an organization's AI projects. I was the only designer in this startup company and led the entire design thinking process from research to launch. Currently, Yennefer is being used by 9 enterprise organizations including Samsung Display, KT Telecom, Hyundai, Seoul National University, and the State University of New York.
Launched T-Mobile GPT
Launched T-Mobile GPT
Launched T-Mobile GPT
Built T-Mobile GPT and it adopted into T-Mobile AI ecosystem
Built T-Mobile GPT and it adopted into T-Mobile AI ecosystem
Built T-Mobile GPT and it adopted into T-Mobile AI ecosystem
PROBLEM
PROBLEM
PROBLEM
Over 65% of customers said switch process was a hassle. Despite additional digital support, 80% of new customers ended up visiting retail stores to switch to T-Mobile.
Over 65% of customers said switch process was a hassle. Despite additional digital support, 80% of new customers ended up visiting retail stores to switch to T-Mobile.
Over 65% of customers said switch process was a hassle. Despite additional digital support, 80% of new customers ended up visiting retail stores to switch to T-Mobile.
RESEARCH
RESEARCH
RESEARCH
Why can’t customers complete the switching journey digitally?
Why can’t customers complete the switching journey digitally?
Why can’t customers complete the switching journey digitally?
🧩
🧩
🧩
Lack of Personalization
Lack of Personalization
Lack of personalized and contextualized reponse
Lack of personalized and contextualized reponse
Lack of personalized and contextualized reponse
📚
📚
📚
Too much Information
Too much Information
Overwhelming information and decision paralysis
Overwhelming information and decision paralysis
Overwhelming information and decision paralysis
📣
📣
📣
Hard to communicate
Hard to communicate
Inefficiency in communication and trust deficiency
Inefficiency in communication and trust deficiency
Inefficiency in communication and trust deficiency
SOLUTION
SOLUTION
SOLUTION
Bringing the in-store Mobile Expert experience to digital
Bringing the in-store Mobile Expert experience to digital
Bringing the in-store Mobile Expert experience to digital
I designed an AI-powered Mobile Expert experience that integrates ChatGPT with T-Mobile’s public product and plan data. The solution replicates the guidance of in-store staff through conversational chat and voice, helping customers confidently complete the switching journey online.
I designed an AI-powered Mobile Expert experience that integrates ChatGPT with T-Mobile’s public product and plan data. The solution replicates the guidance of in-store staff through conversational chat and voice, helping customers confidently complete the switching journey online.
I designed an AI-powered Mobile Expert experience that integrates ChatGPT with T-Mobile’s public product and plan data. The solution replicates the guidance of in-store staff through conversational chat and voice, helping customers confidently complete the switching journey online.
DATA-DRIVEN DECISION
DATA-DRIVEN DECISION
DATA-DRIVEN DECISION
Why an agentic AI chatbot?
Why an agentic AI chatbot?
Why an agentic AI chatbot?
I led ideation workshops that generated 100+ ideas and helped prioritize the final direction using user and business insights.
I led ideation workshops that generated 100+ ideas and helped prioritize the final direction using user and business insights.
I led ideation workshops that generated 100+ ideas and helped prioritize the final direction using user and business insights.
x
x
x
Chat drives conversion
Chat drives conversion
Customers who used chat showed a 2.5× higher conversion rate, even when the experience was imperfect.
Customers who used chat showed a 2.5× higher conversion rate, even when the experience was imperfect.
Customers who used chat showed a 2.5× higher conversion rate, even when the experience was imperfect.
%
%
%
Prefer conversational support
Prefer conversational support
survey respondents reported positive emotions when interacting with chat.
survey respondents reported positive emotions when interacting with chat.
survey respondents reported positive emotions when interacting with chat.
%
%
%
But, existing chatbot fails to complete
But, existing chatbot fails to complete
But, existing chatbot fails to complete
users needed additional steps after chat because the bot didn’t fully understand their needs or provide clear answers.
users needed additional steps after chat because the bot didn’t fully understand their needs or provide clear answers.
users needed additional steps after chat because the bot didn’t fully understand their needs or provide clear answers.
Design
Design
Design
PROCESS
PROCESS
End-to-end design in a startup-like environment
End-to-end design in a startup-like environment
End-to-end design in a startup-like environment
To evaluate our idea quickly through an MVP test, I defined key scenarios and features with PM. We had 3 weeks sprint to build a fully working demo application, including design, development, and QA test.
To evaluate our idea quickly through an MVP test, I defined key scenarios and features with PM. We had 3 weeks sprint to build a fully working demo application, including design, development, and QA test.
To evaluate our idea quickly through an MVP test, I defined key scenarios and features with PM. We had 3 weeks sprint to build a fully working demo application, including design, development, and QA test.
Wireframes
Wireframes
Wireframes
What’s the first image in our minds
What’s the first image in our minds
What’s the first image in our minds
AI engineering
AI Engineering
AI engineering
What’s the first image in our minds
What’s the first image in our minds
What’s the first image in our minds
MVP Test
MVP Test
MVP Test
What’s the first image in our minds
What’s the first image in our minds
What’s the first image in our minds
CHALLENGE
CHALLENGE
CHALLENGE
Good UX vs Speed
Good UX vs Speed
Good UX vs Speed
Since we had limited time to create a fully working demo application, I had to find the balance between implementation effort and user experiences. I worked closely with engineers and pm to discuss feasibility and priorities.
Since we had limited time to create a fully working demo application, I had to find the balance between implementation effort and user experiences. I worked closely with engineers and pm to discuss feasibility and priorities.
Since we had limited time to create a fully working demo application, I had to find the balance between implementation effort and user experiences. I worked closely with engineers and pm to discuss feasibility and priorities.
A Horizontal scroll
A Horizontal scroll
B Accordion
B Accordion
C Direct PDP link
C Direct PDP link
MVP Test
MVP Test
IMPROVEMENT 1
IMPROVEMENT 1
Before
Before
Too many follow up questions 🧐
Too many follow up questions 🧐
Too many follow up questions 🧐
AI requires data to provide personalized recommendations
AI requires data to provide personalized recommendations
AI requires data to provide personalized recommendations
After
After
Replaced to quiz format
Replaced to quiz format
Replaced to quiz format
Users reacted better to the quiz format to response
Users reacted better to the quiz format to response
Users reacted better to the quiz format to response
IMPROVEMENT 2
Before
Before
Not existing design system
Not existing design system
Not existing design system
Since AI chatbot is new product at T-mobile, some components are not existed
Since AI chatbot is new product at T-mobile, some components are not existed
Since AI chatbot is new product at T-mobile, some components are not existed
After
After
After
Created design system components
Created design system components
Created design system components
I created components that align with T-Mobile brading
I created components that align with T-Mobile brading
I created components that align with T-Mobile brading
Design challenge
Design challenge
Design challenge
Challenge 1: How can we obtain data for personalized recommendations without disrupting the chat experience?
Challenge 1: How can we obtain data for personalized recommendations without disrupting the chat experience?
How can we obtain data for personalized recommendations without disrupting the chat experience?
Option A:
Onboarding before chat start
Option A:
Onboarding before chat start
Option B:
Integrate personalized quiz in chat experiece
Option B:
Integrate personalized quiz in chat experiece
1st Solution
Landing page
Poor visualization
1st Solution
Landing page
1st Solution
Landing page
Poor visualization
2nd Solution
Accessible AI
Concise and direct information for questions
2nd Solution
Accessible AI
Concise and direct information for questions
2nd Solution
Accessible AI
Concise and direct information for questions
3rd Solution
Personalized recommendations
Personalized recommendations
3rd Solution
Personalized recommendations
3rd Solution
Personalized recommendations
Personalized recommendations
4th Solution
Voice interaction
Chat and voice interaction with agentic AI
4th Solution
Voice interaction
Chat and voice interaction with agentic AI
4th Solution
Voice interaction
Chat and voice interaction with agentic AI
IMPACT
IMPACT
IMPACT
MVP Test KPIs
MVP Test KPIs
MVP Test KPIs
%
%
%
Task success rate
Task success rate
Task success rate
Unlike T-Mobile current platform, users completed their tasks without support
Unlike T-Mobile current platform, users completed their tasks without support
Unlike T-Mobile current platform, users completed their tasks without support
%
%
%
Faster Average interaction duration
Faster Average interaction duration
Faster Average interaction duration
Task completion and decision making time reduced from 12.5m to 2.5m
Task completion and decision making time reduced from 12.5m to 2.5m
Task completion and decision making time reduced from 12.5m to 2.5m
AI Patents
AI Patents
AI Patents
Digital Innovation Lab’s success metric is make conceptual idea to production
Digital Innovation Lab’s success metric is make conceptual idea to production
Digital Innovation Lab’s success metric is make conceptual idea to production
