Chatlighter
Save the useful moment, not the whole conversation
OVERVIEW
I designed Chatlighter, a Chrome extension that lets users highlight and save specific moments from AI conversations instead of keeping entire threads. The experience is designed around quick capture first, with lightweight organization available when users need it later.
Time Aug 2026 (Independent Project)
Role Product Design · Interaction Design · Prototyping
Tools Figma, HTML/CSS/JavaScript, Chrome Extension API, AI-assisted coding
PROBLEM
The useful part is often smaller than the conversation.
AI conversations often grow through follow-up questions, revisions, and topic changes. Users may only need one specific recommendation, explanation, or decision, but returning to it often means reopening the full conversation and manually searching through the thread again. How might we help users capture and return to the exact moment that matters?
RESEARCH
People save information, but struggle to return to the right moment.
I surveyed 10 frequent AI users to understand how they revisit and save useful information from past conversations.
9/10 frequently switch topics within the same AI conversation
9/10 return to previous conversations to find specific information
60% use screenshots to save useful responses
These findings suggested that users do not always need to save an entire conversation. They need a smaller, more retrievable unit of information.
Survey conducted in Korean and English. Korean responses were translated into English for this case study. 5–10 min · Online survey (Multiple-choice + open-ended questions)
DESIGNING THE MOMENT
Capture should feel like part of reading.
Instead of introducing a separate saving workflow, I designed the interaction around a familiar behavior: selecting text.
Users can highlight the exact part of an AI response they want to keep and save it directly from the conversation.
The goal was to make saving feel lightweight and immediate without interrupting the flow of reading.
HOW IT WORKS
Highlight, Save, Organize, and Return.
01 HIGHLIGHT
Users should be able to save a specific sentence, paragraph, or section instead of an entire conversation. Every saved highlight should maintain a connection to the conversation it came from.
02 SAVE & ORGANIZE
Saving should happen quickly without forcing users to organize information in the moment. Saved highlights can later be grouped into projects as the collection grows.
04 RETURN
Each highlight keeps a link back to the original conversation so users can recover the surrounding context when needed.
PROTOTYPE & ITERATION
Testing the interaction inside a real browser.
I moved from Figma to a functional Chrome extension to evaluate the interaction in a real browser environment. This helped me understand how text selection, save feedback, and the extension itself would behave within an existing AI interface.
One key design decision was to prioritize capture over organization. Users can save a useful moment immediately, then organize it later, keeping the interaction fast and lightweight.
I used AI coding tools to generate and revise the implementation based on the interaction and behavior I designed. I then tested the extension in the browser, identified usability and interaction issues, and refined the experience through iteration.
REFLECTION
A small interaction can solve a focused problem.
Chatlighter helped me see that saving information is only useful if it is easy to return to later. Instead of expanding the product with more features, I focused on making one simple action—saving a useful part of a conversation—quick and easy within the existing chat experience.
If I continued the project, I would explore simple ways to search and organize saved highlights as the collection grows.