*Please note that due to NDA restrictions of upcoming unreleased features, the full experience cannot be shown here. Please contact me for a complete demonstration.

HEYCONTENT

Designing an AI system that learns from creators and turns scattered ideas into consistent progress.

jump to prototype

My role

Product Designer

Team

Designers, Engineer, Marketing

timeline

3 months (2025)

✧ What if content creation could learn and grow with you instead of resetting every session?

Most AI tools generate outputs quickly but fail to retain context or reflect progress over time.

HeyContent introduces an adaptive system that builds context, transforming one-time interactions into a continuous workflow. By understanding user intent, capturing knowledge into structured memory, and reflecting progress through feedback, the system helps creators refine direction and build momentum instead of starting from scratch.

HeyContent introduces an adaptive system that builds context, transforming one-time interactions into a continuous workflow.
By understanding user intent, capturing knowledge into structured memory, and reflecting progress through feedback, the system helps creators refine direction and build momentum instead of starting from scratch.

✧ My Role

As a Product Designer, I led end-to-end product design from concept exploration to prototype validation.

I worked closely with engineers and stakeholders to translate creator needs into structured workflows from research synthesis and interaction design to usability testing and iteration.

I drove key decisions around system behavior, user feedback patterns, and feature prioritization, ensuring the experience balanced AI intelligence with clarity, trust, and user control.

I worked closely with engineers and stakeholders to translate creator needs into structured workflows from research synthesis and interaction design to usability testing and iteration.

I drove key decisions around system behavior, user feedback patterns, and feature prioritization, ensuring the experience balanced AI intelligence with clarity, trust, and user control.

I worked closely with engineers and stakeholders to translate creator needs into structured workflows from research synthesis and interaction design to usability testing and iteration.

I drove key decisions around system behavior, user feedback patterns, and feature prioritization, ensuring the experience balanced AI intelligence with clarity, trust, and user control.

✧ Problem

Creators juggle disconnected tools and repeatedly rebuild context, which breaks momentum.

They struggle to capture ideas quickly, track progress over time, and receive guidance that adapts to their style. Because context does not persist across tools or sessions, creators lose direction, feel overwhelmed, and struggle to build consistent growth.

✧ Core Insights

The issue isn’t generating content. It’s maintaining continuity.

Interviews with 3–5 creators revealed frustration with fragmented tools, difficulty staying organized, and AI systems that reset context every session. Participants wanted personalized guidance, structured workflows, and a clearer sense of progress over time, which reframed the problem from producing outputs to supporting long-term momentum.

✧ System Overview

I structured HeyContent around three connected capabilities: understanding, memory, and growth.

The experience is structured as a continuous learning loop. Conversations help the system understand intent, insights are captured into persistent memory, and progress is reflected back through feedback over time. These capabilities surface through three core experiences — chat, Smart Notes workspace, and the timeline — creating continuity across the creator’s workflow.

✧ Goals

Design a workflow that feels simple in the moment but becomes smarter over time.

The experience needed to support discovering ideas, organizing them into projects, and tracking evolution without adding complexity. I focused on making system state visible, reducing uncertainty, and keeping users in control while interacting with AI.

✧ How Might We

How might we create an AI co-pilot that grows with creators and turns ideas into action within one continuous workflow?

✧ Design

I explored multiple ways to connect conversation, organization, and progress into one system.

I explored multiple ways to connect conversation, organization, and progress into
one system.

I explored multiple ways to connect conversation, organization, and progress into one system.

Early concepts tested how context should move between chat and workspace, how memory should be structured, and how progress could be visualized without overwhelming users. These explorations informed the direction of the V1 prototype.

Creators shared that their tools are scattered, hard to manage, and don’t adapt to their goals or style. They want a simple, personalized workspace that helps them stay consistent and organized. These insights informed the core features of HeyContent, from persona setup to timeline tracking and Smart Notes.

✴︎ Capability 1: AI Understanding

Chat acts as the entry point where the system interprets user intent and begins building contextual understanding.

Designed a guided persona setup flow that supports multiple inputs and provides clear feedback during interaction. The improvements focused on making progress visible, clarifying system status, and allowing users to update their persona easily after setup, reducing uncertainty while helping the system learn more accurate preferences over time.

✴︎ Capability 2: AI Memory

Smart Notes transform temporary conversations into persistent, organized knowledge.

Smart Notes transform temporary conversations into persistent organized knowledge.

Smart Notes transform temporary conversations into persistent, organized knowledge.

I designed both embedded capture and a full workspace that allows users to organize notes into projects and folders. This supports continuity by ensuring valuable ideas remain accessible and reusable over time.

I designed both embedded capture and a full workspace that allows users to organize notes into projects and folders.
This supports continuity by ensuring valuable ideas remain accessible and reusable over time.

Embedded note panel

Full workspace

✴︎ Capability 3: AI Growth

The timeline makes progress visible so creators can understand how they evolve over time.

Users wanted a quick way to recognize milestones and patterns across their activity. The timeline connects actions to outcomes, reinforcing motivation and helping creators make more informed decisions about their strategy.

✧ V1 Prototype - Usability Testing

Testing focused on whether users understood system feedback and could complete tasks confidently.

I observed how creators interacted with persona setup and AI prompts, identifying confusion around progress visibility and editing controls. Insights directly informed design improvements that increased clarity and confidence.

Key Observations ⤸

Unclear System Status During Persona Creation

Users are unsure when the persona icon appears, whether it updates in real time, and how long they need to keep interacting with the AI before their persona is summarized.

Lack of Editable Feedback After Generation

Users want a clear and easy way to update or correct their persona after viewing it, highlighting a gap in editable interactions.

Single-Choice Inputs Don’t Reflect Creator Reality

Users may want to select more than one answer or layer responses (e.g., goals, content formats, platforms). The current one-answer flow feels restrictive and doesn’t reflect creators’ reality, highlighting the need for multi-select or expandable input options.

Quote

“I’m not sure when my persona is ready or if it’s updating. How long do I need to keep talking to the AI until it is ready?”

✧ From Insight to Iteration

Improvements focused on increasing clarity, control, and confidence in AI interactions.

Usability testing revealed confusion around system status, editability, and input flexibility, particularly during persona setup in the chat experience. I translated these insights into targeted design changes, prioritizing clearer feedback, more predictable interaction patterns, and greater user control at critical moments in the workflow.

Improve Persona Visibility & Setup Feedback

Make it clear when the persona is active, how it updates in real time, and how long setup takes, so users always know their persona status and progress.

Simplify Editing & Updates

Provide an intuitive way for users to edit or correct their persona after setup, ensuring the process is seamless and low-friction.

Supporting Multiple Inputs & Flexible Responses

Allow users to select or layer multiple goals, platforms, and content formats, reflecting the reality of creators’ workflows and avoiding restrictive one-answer flows.

These improvements focused on the chat persona flow as the system’s entry point, where early confusion had the greatest impact on user confidence.

Before

After

Horizontal progress bar

To guide users through each step of persona setup, reducing uncertainty about completion.

Multiple Inputs

Users can now select multiple inputs via checkboxes, allowing them to reflect answers.

Clear Visual Feedback

The persona button remains grayed out until setup is complete, giving clear visual feedback.

✧ Final Prototype

Capability 1: AI Understanding
Interprets user intent and begins building contextual understanding.
Capability 2: AI Memory
Transform temporary conversations into persistent, organized knowledge.
Capability 3: AI Growth
Makes progress visible so creators can understand how they evolve over time.

Capability 1: AI Understanding
Interprets user intent and begins building contextual understanding.
Capability 2: AI Memory
Transform temporary conversations into persistent, organized knowledge.
Capability 3: AI Growth
Makes progress visible so creators can understand how they evolve over time.

✧ Impact Summary

The redesigned flows improved clarity and established a scalable foundation for future features.

The redesigned flows improved clarity and established a scalable foundation for
future features.

The redesigned flows improved clarity and established a scalable foundation for future features.

Updates reduced onboarding confusion and introduced reusable patterns for multi-input responses and progress indicators. The work informed product direction by defining interaction principles for AI-driven workflows.

Updates reduced onboarding confusion and introduced reusable patterns for multi-input responses and progress indicators.
The work informed product direction by defining interaction principles for AI-driven workflows.

✧ Next Step

Next steps focus on validating long-term value and optimizing the workflow loop.

Test First-Time Setup Flow

Observe how new users respond to the grayed-out persona button to ensure it communicates clearly that setup must be completed before editing.

Evaluate In-Context Editing

Gather feedback on the new Edit option inside the persona popup to confirm it reduces confusion and prevents unnecessary back-and-forth navigation.

Monitor Onboarding Experience

Measure whether these changes improve overall clarity and ease of setting up a persona during the first interaction with the AI.

Measure whether these changes improve overall clarity and ease of setting up a persona during the first interaction
with the AI.

Measure whether these changes improve overall clarity and ease of setting up a persona during the first interaction with the AI.

Capability 1: AI Understanding
Interprets user intent and begins building contextual understanding.
Capability 2: AI Memory
Transform temporary conversations into persistent, organized knowledge.
Capability 3: AI Growth
Makes progress visible so creators can understand how they evolve over time.

✧ Final Prototype

Capability 1: AI Understanding
Interprets user intent and begins building contextual understanding.
Capability 2: AI Memory
Transform temporary conversations into persistent, organized knowledge.
Capability 3: AI Growth
Makes progress visible so creators can understand how they evolve over time.

✧ Final Prototype