Overview
As Senior Product Designer and Design Systems Lead, I led the creation of Ask Intel, the company’s first unified conversational AI platform.
Rather than launching another support widget, I architected an AI Assistant Layer that now serves as a scalable orchestration system across support, documentation and contact flows.
This was not a UI project. It was a systems redesign.
Role & Scope
I led the end-to-end design of Ask Intel, the company's first enterprise-wide conversational assistant. This meant defining the conversational model, building the chat UI and creating a scalable component system. This was a ground-up effort. No prior assistant, no existing framework, no shared model for conversational behavior. I established the foundation that Intel's conversational experiences continue to scale on.
Challenge
Intel's legacy chat tools were:
Each group optimized locally. Users experienced fragmentation globally.
The mandate was clear: Move from “chat as add-on” to AI as a primary navigation and resolution layer.
Accessibility as Architecture
Conversational interfaces break differently than pages. I engineered:
Testing with keyboard-only and non-visual users surfaced logic gaps earlier than any prototype review.
I led a custom rebuild:
Key Insights
Impact
My Approach: Build the Foundation First
We aligned taxonomy, metadata and content governance before scaling. Before designing conversations, I focused on infrastructure.
The Assistant Layer
Conversational AI is systems design, not just UI. Success required aligning taxonomy, content quality, NLP capabilities, localization and user expectations. Not just building a chat window.
The Experience Layer
Enterprise users demand clarity and trust. Conversational design must be transparent, concise and predictable to work effectively in large organizations. Vague or uncertain responses erode confidence quickly.
The Explainer Layer
Machine-readable, modular content blocks designed for both internal logic and external LLM citation.
Detailed Results
When systems break, teams slow down.
I work across UX, architecture and content to prevent fragmentation and help organizations move faster with confidence.
Overview
As Senior Product Designer and Design Systems Lead, I led the creation of Ask Intel, the company’s first unified conversational AI platform.
Rather than launching another support widget, I architected an AI Assistant Layer that now serves as a scalable orchestration system across support, documentation and contact flows.
This was not a UI project. It was a systems redesign.
Challenge
Intel's legacy chat tools were:
Each group optimized locally. Users experienced fragmentation globally.
The mandate was clear: Move from “chat as add-on” to AI as a primary navigation and resolution layer.
Accessibility as Architecture
Conversational interfaces break differently than pages. I engineered:
Testing with keyboard-only and non-visual users surfaced logic gaps earlier than any prototype review.
I led a custom rebuild:
Impact
My Approach: Build the Foundation First
We aligned taxonomy, metadata and content governance before scaling. Before designing conversations, I focused on infrastructure.
The Assistant Layer
Conversational AI is systems design, not just UI. Success required aligning taxonomy, content quality, NLP capabilities, localization and user expectations. Not just building a chat window.
The Experience Layer
Enterprise users demand clarity and trust. Conversational design must be transparent, concise and predictable to work effectively in large organizations. Vague or uncertain responses erode confidence quickly.
The Explainer Layer
Machine-readable, modular content blocks designed for both internal logic and external LLM citation.
Detailed Results
Overview
As Senior Product Designer and Design Systems Lead, I led the creation of Ask Intel, the company’s first unified conversational AI platform.
Rather than launching another support widget, I architected an AI Assistant Layer that now serves as a scalable orchestration system across support, documentation and contact flows.
This was not a UI project. It was a systems redesign.
Challenge
Intel's legacy chat tools were:
Each group optimized locally. Users experienced fragmentation globally.
The mandate was clear: Move from “chat as add-on” to AI as a primary navigation and resolution layer.
Accessibility as Architecture
Conversational interfaces break differently than pages. I engineered:
Testing with keyboard-only and non-visual users surfaced logic gaps earlier than any prototype review.
I led a custom rebuild:
Impact
My Approach: Build the Foundation First
We aligned taxonomy, metadata and content governance before scaling. Before designing conversations, I focused on infrastructure.
The Assistant Layer
Conversational AI is systems design, not just UI. Success required aligning taxonomy, content quality, NLP capabilities, localization and user expectations. Not just building a chat window.
The Experience Layer
Enterprise users demand clarity and trust. Conversational design must be transparent, concise and predictable to work effectively in large organizations. Vague or uncertain responses erode confidence quickly.
The Explainer Layer
Machine-readable, modular content blocks designed for both internal logic and external LLM citation.
Detailed Results
Overview
As Senior Product Designer and Design Systems Lead, I led the creation of Ask Intel, the company’s first unified conversational AI platform.
Rather than launching another support widget, I architected an AI Assistant Layer that now serves as a scalable orchestration system across support, documentation and contact flows.
This was not a UI project. It was a systems redesign.
Challenge
Intel's legacy chat tools were:
Each group optimized locally. Users experienced fragmentation globally.
The mandate was clear: Move from “chat as add-on” to AI as a primary navigation and resolution layer.
Accessibility as Architecture
Conversational interfaces break differently than pages. I engineered:
Testing with keyboard-only and non-visual users surfaced logic gaps earlier than any prototype review.
I led a custom rebuild:
Impact
KPI
Support Deflection
Improvement
22%
Metric Type
Operational Efficiency
KPI
A11y Speed
Improvement
35% faster
Metric Type
Accessibility/CX Speed
KPI
Task Success
Improvement
92%
Metric Type
User Proficiency
KPI
Compliance
Improvement
100% WCAG 2.1 AA
Metric Type
Technical Standard
My Approach: Build the Foundation First
We aligned taxonomy, metadata and content governance before scaling. Before designing conversations, I focused on infrastructure.
The Assistant Layer
Conversational AI is systems design, not just UI. Success required aligning taxonomy, content quality, NLP capabilities, localization and user expectations. Not just building a chat window.
The Experience Layer
Enterprise users demand clarity and trust. Conversational design must be transparent, concise and predictable to work effectively in large organizations. Vague or uncertain responses erode confidence quickly.
The Explainer Layer
Machine-readable, modular content blocks designed for both internal logic and external LLM citation.
Detailed Results
Overview
As Senior Product Designer and Design Systems Lead, I led the creation of Ask Intel, the company’s first unified conversational AI platform.
Rather than launching another support widget, I architected an AI Assistant Layer that now serves as a scalable orchestration system across support, documentation and contact flows.
This was not a UI project. It was a systems redesign.
Challenge
Intel's legacy chat tools were:
Each group optimized locally. Users experienced fragmentation globally.
The mandate was clear: Move from “chat as add-on” to AI as a primary navigation and resolution layer.
Accessibility as Architecture
Conversational interfaces break differently than pages. I engineered:
Testing with keyboard-only and non-visual users surfaced logic gaps earlier than any prototype review.
I led a custom rebuild:
Impact
KPI
Support Deflection
Improvement
22%
Metric Type
Operational Efficiency
KPI
A11y Speed
Improvement
35% faster
Metric Type
Accessibility/CX Speed
KPI
Task Success
Improvement
92%
Metric Type
User Proficiency
KPI
Compliance
Improvement
100% WCAG 2.1 AA
Metric Type
Technical Standard
My Approach: Build the Foundation First
We aligned taxonomy, metadata and content governance before scaling. Before designing conversations, I focused on infrastructure.
The Assistant Layer
Conversational AI is systems design, not just UI. Success required aligning taxonomy, content quality, NLP capabilities, localization and user expectations. Not just building a chat window.
The Experience Layer
Enterprise users demand clarity and trust. Conversational design must be transparent, concise and predictable to work effectively in large organizations. Vague or uncertain responses erode confidence quickly.
The Explainer Layer
Machine-readable, modular content blocks designed for both internal logic and external LLM citation.
Detailed Results
Role & Scope
I led the end-to-end design of Ask Intel, the company's first enterprise-wide conversational assistant. This meant defining the conversational model, building the chat UI and creating a scalable component system. This was a ground-up effort. No prior assistant, no existing framework, no shared model for conversational behavior. I established the foundation that Intel's conversational experiences continue to scale on.
Role & Scope
I led the end-to-end design of Ask Intel, the company's first enterprise-wide conversational assistant. This meant defining the conversational model, building the chat UI and creating a scalable component system. This was a ground-up effort. No prior assistant, no existing framework, no shared model for conversational behavior. I established the foundation that Intel's conversational experiences continue to scale on.
Role & Scope
I led the end-to-end design of Ask Intel, the company's first enterprise-wide conversational assistant. This meant defining the conversational model, building the chat UI and creating a scalable component system. This was a ground-up effort. No prior assistant, no existing framework, no shared model for conversational behavior. I established the foundation that Intel's conversational experiences continue to scale on.
Role & Scope
I led the end-to-end design of Ask Intel, the company's first enterprise-wide conversational assistant. This meant defining the conversational model, building the chat UI and creating a scalable component system. This was a ground-up effort. No prior assistant, no existing framework, no shared model for conversational behavior. I established the foundation that Intel's conversational experiences continue to scale on.