Work

About

Resume

Flow

Master’s Capstone project sponsored by Microsoft.

I led the end-to-end design of Flow, a neurodivergent-friendly AI assistant and toolkit, that supports neurodivergent workers through its 3 key features: Magic Mouse, Activity History, and a Custom AI System Prompt.

Flow reduces context switching, eases communication stress, and scaffolds focus - transforming AI from a passive tool into a proactive partner, setting a new direction for inclusive, future-forward productivity design.

Sponsor

Microsoft

Timeline

8 Months (Jan - Aug 2025)

Role

Product Designer

Team

1 Principal PM, Microsoft

1 Senior UX Researcher, Microsoft

4 Designers

Impact

Time Saved

25 mins

Users context switch less and stay in flow for longer, saving a projected 25 mins per day.

AI Prompt Satisfaction

4.3/5

25 surveyed participants rated their satisfaction with the AI system prompt an average of 4.3 out of 5.

Adoption & Usability

100%

All 4 usability test participants stated that Flow would improve their workflow, and expressed interest in adopting it.

Microsoft's Recognition

"This is such a real problem affecting many AI applications today. Love the focus on reducing cognitive load and it following you wherever you go! This is really cool to see!"

— Senior PM @ Microsoft

"Wow, I'm looking at this and I feel like I want something like this. When I talk to my manager, this would be super helpful for me personally."

— Senior Product Designer @ Microsoft

Problem Discovery

ADHD Professionals Face Stigma and Strain

Microsoft's Neurodivergent Employee Resource Group highlighted how ADHD challenges are amplified in their high-demand tech roles, and stigma often prevents them from seeking help. Microsoft challenged us to consider how AI can intervene and empower neurodivergent employees.

Scope

Focused on neurodivergent tech workers (Product Managers, Software Engineers and Designers) who experience high-demand environments that amplify ADHD pain points.

Scalability & Business Impact

If our concept worked in this context, it could scale to other knowledge workers, aligned with Microsoft’s goals of supporting all employees and shaping productivity tools.

Insights Informed by Lived Experiences & ADHD Specialists

To understand the problem space, I conducted desk and generative research. To respect the sensitive and medical nature of ADHD, I created a study plan, recruited and conducted a mixed-methods study to learn from both lived experiences and ADHD specialists (including a psychiatrist).

—> open Generative Research Report in new tab

Insight 1: Navigating Social Communication

Insight 2: Supporting Shifting Focus & Energy

Insight 3: Minimizing Disruptions & Context Loss

Insight 1: Navigating Social Communication

Insight 2: Supporting Shifting Focus & Energy

Insight 3: Minimizing Disruptions & Context Loss

How Might We

Our research revealed a powerful mindset shift: participants emphasized that they did not view ADHD as a limitation, but as a source of superpowers like hyperfocus, non-linear thinking, and authenticity. Yet, recurring pain points like context switching, overwhelm, and workplace communication stress, often keep these superpowers from shining.

The challenge: How might we amplify these strengths while reducing the cognitive and emotional friction that holds them back?

How Might We

The challenge: How might we amplify these strengths while reducing the cognitive and emotional friction that holds them back?

Final Design

Magic Mouse

Magic Mouse reduces context switching and cognitive load by letting users capture, query, and act on information directly in their workflow: no copy-paste or app-switching required. It also dynamically shifts between "text highlight" or "area select" mode for adaptive, context-aware support.

Magic Mouse: Text Highlight mode

Where longer, multi-page text or precision is required.

Magic Mouse: Area Select mode

Where non-text assets (e.g. images) are involved.

ADHD-Friendly System Prompt

I led the design of a custom system prompt of our AI Assistant. This custom prompt turns generic responses into a proactive, supportive partner. It is engineered to provide emotional scaffolding, anticipate overwhelm, offer simple next steps, and communicate with empathy and clarity.

—> open Flow AI Assistant in new tab

Activity History

Activity History passively captures users work sessions through screen recording. It then creates a dynamic, searchable timeline that helps users quickly revisit and recover from interruptions, sustaining their flow and momentum.

Design Overview

2-Layered Approach: UI & AI Layers

Flow needed to be designed at two layers because neurodivergent pain points show up both in how users access support and in how support is delivered.

Design Guidelines

I created design guidelines applicable to both interface and prompt design.

Be Quietly Available, Never Intrusive

Proactive & Frictionless

Emotionally Safe

Adaptive & Personalized

Interface Design

UI Layer

We ideated widely, creating and refining user personas and information architecture. Through rapid sprints, design critiques with Microsoft Designers and University Professors, and user testing, we converged on a design.

How Many Features?

Initially, we wanted to accommodate every possible need that users may have. However, user testing and feedback revealed that less is more. Focusing on the features users valued most, enabled us to avoid repetition, streamline the design and optimize each feature's impact.

Toolbar Function & Placement: Balancing Visibility With Simplicity

The toolbar serves as Flow’s entry point for users to access its features and the AI chat. I experimented with different placements and levels of progressive disclosure to balance 2 competing needs: providing enough context for users to understand Flow's features, while keeping the toolbar lightweight and non-intrusive.

Early iterations tested expandable toolbars which surfaced features either one at a time or in a summary view. While this offered flexibility, I realized that progressive disclosure introduced additional cognitive load: users had to click, expand, and scan before accessing what they needed. The final version is minimal and non-collapsible, centered at the bottom of the screen. Its consistent placement made it easy to find, while its compact size kept the workspace clear. This made it feel more like a supportive layer rather than a dominant application, accessible but not overwhelming.

Iteration 1

Expandable toolbar at the top-right corner of screen to keep it easily accessible yet unobtrusive. In the expanded view, only 1 feature appeared at a time to reduce visual clutter.

Iteration 2

Expandable toolbar at bottom-right of screen with a summary view of all features at a glance. This placement aligned with desktop interaction patterns and provided more access to information.

Final Version

Fixed toolbar centered at the bottom, allowing it to be quickly accessed and non-obtrusive, while keeping cognitive load to a minimum.

OS layer vs. App

We first explored an OS-layer integration, which would have allowed Flow to run seamlessly in the background like a built-in feature. This approach is frictionless: no need to download a separate app, manually connect tools, or repeatedly grant access permissions.

However, an OS-layer risks excluding users on older systems, and raises complexity around privacy, updates, and cross-platform compatibility. By positioning Flow as a Microsoft Suite app instead, users could enable it with the same access permissions and authentication they already trust for Word, Teams, and Outlook. This lowers adoption barriers and makes integration smoother, more secure, and scalable across both Windows and non-Windows environments.

Iteration 1

Expandable toolbar at the top-right corner of screen to keep it easily accessible yet unobtrusive. In the expanded view, only 1 feature appeared at a time to reduce visual clutter.

Iteration 2

Expandable toolbar at bottom-right of screen with a summary view of all features at a glance. This placement aligned with desktop interaction patterns and provided more access to information.

Final Version

Minimal toolbar centered at the bottom edge. This layout strikes the balance of being quickly accessed, not visually-distracting and usable without disrupting focus.

Prompt Design

AI Layer

Over half (56.4%) of participants described workplace communication as one of their biggest pain points; often avoiding replies when emotionally overloaded. This revealed a need to go beyond interface design and introduce an AI prompt layer to scaffold communication through tailored, emotionally-aware responses. I led the design of Flow's interaction blueprint, designed on LLM GPT-4o, deployed via Microsoft Azure AI Foundry.

Tested & Approved By Users, Psychiatrist & ADHD Coach

Round 1: Internal Evaluation

Round 2: ADHD User Evaluation

Round 3: Specialist Evaluation

How It Differs From Standard GPT?

Comparing the response to the same input, our customGPT is proactive, supportive and low-friction. Designed for neurodivergent users, it prioritizes psychological safety, empathy and actionable steps.

Standard GPT-4o

Flow's Custom GPT

Hand Off

Stakeholder Presentation

To hand off, we gave a final presentation to our sponsor and Master’s program professors. We also delivered design artifacts, including high-fidelity prototypes, the system prompt, and research documentation.

Next Steps

Our Microsoft sponsor told us that they will adapt and implement Flow’s designs, using our deliverables as a foundation for future iterations.

Impact Recap