Web app
Productivity
Product Design
2024
How I designed an AI companion feature to end dull
tasks
Researching, UX/UI, Design Handoff
Background
Lawyers who used Lupl needed ways to get more done in less time. I made an AI feature that helped lawyers get their work done faster.
Impact
MVP launched within 2 months
Helped with expanding the design system into a base that can be used by anyone
Made it easier for users to create matters and get useful insights
Problem: unreliable AI output
One major problem was that AI didn't always give the same answer, which could make it hard for prototype designs to work in real life. Another problem was finding the right balance between using contextual AI and handling auto-enabling correctly.
Design explorations
After exploring various design ideas, I reached a point where I felt confident enough. Ready to get feedback on the first designs.
First try:
Prebaked prompts for welcome modal
Context-aware AI for collaborators
Context-aware AI for notifications
Prebuilt prompts for text
Getting feedback
During a feedback session, we concluded that the color I picked didn’t match the brand. It had to stand out, but the above purple was too different from Lupl’s teal identity.
A key takeaway from this meeting was to use familiarity. Since our customers were already using Microsoft Copilot, we decided on this card system for our assistant, as shown below.
Focus change
Although the first designs were inspired by Notion's context-aware AI assistant, things turned out to be different along the way. Due to a development focus switch, the new requirements for a minimum viable product had been much simplified.
The new needs included four types of prompts on the dashboard AI modal. These cards had buttons that let you quickly turn text into tasks in Lupl.
Improvement: Instead of displaying text prompts one by one, they were arranged in cards to make it easier for users to choose.
AI output was simplified to make development easier and create a basic version of the product to test.
Matter-related cards were simplified for the final iteration. They provide quick follow-up
actions for quickly creating items.
Task-related cards are simple and display only the essential information and a progress status.
Lessons learned
Unfortunately, in the end, this feature was completely deprioritized. But the iterative design process helped me gain valuable insights into meeting technical and business constraints.
It helped me develop an agile attitude and become more aware that not all design ideas will make it into production. And most importantly, it gave me an opportunity to learn and improve.
Web app
Productivity
Product Design
2024
How I designed an AI companion feature to end dull tasks
Researching, UX/UI, Design Handoff
Background
Lupl is a tool designed to streamline legal processes and workflows. I build a step feature that turns overload into manageable and enables users to better manage their workload.
Impact
MVP launched within 2 months
Helped with expanding the design system into a base that can be used by anyone
Made it easier for users to create matters and get useful insights
Problem: unreliable AI output
One major problem was that AI didn't always give the same answer, which could make it hard for prototype designs to work in real life. Another problem was finding the right balance between using contextual AI and handling auto-enabling correctly.
Design explorations
After exploring various design ideas, I reached a point where I felt confident enough. Ready to get feedback on the first designs.
First try: Prebaked prompts for welcome modal
Context-aware AI for collaborators
Context-aware AI for notifications
Prebuilt prompts for text
Getting feedback
During a feedback session, we concluded that the color I picked didn’t match the brand. It had to stand out, but the above purple was too different from Lupl’s teal identity.
A key takeaway from this meeting was to use familiarity. Since our customers were already using Microsoft Copilot, we decided on this card system for our assistant, as shown below.
Focus change
Although the first designs were inspired by Notion's context-aware AI assistant, things turned out to be different along the way. Due to a development focus switch, the new requirements for a minimum viable product had been much simplified.
The new needs included four types of prompts on the dashboard AI modal. These cards had buttons that let you quickly turn text into tasks in Lupl.
Improvement: Instead of displaying text prompts one by one, they were arranged in cards to make it easier for users to choose.
Matter-related cards were simplified for the final iteration. They provide quick follow-up actions for quickly creating items.
Task-related cards are simple and display only the essential information and a progress status.
Additional improvements
Unfortunately, in the end, this feature was completely deprioritized. But the iterative design process helped me gain valuable insights into meeting technical and business constraints.
It helped me develop an agile attitude and become more aware that not all design ideas will make it into production. And most importantly, it gave me an opportunity to learn and improve.
Web app
Productivity
Product Design
2024
How I designed an AI companion feature to end dull tasks
Researching, UX/UI, Design Handoff
Background
I improved the checkout page to increase donations for Declic - a non-profit group for online petitions and activism.
Impact
MVP launched within 2 months
Helped with expanding the design system into a base that can be used by anyone
Made it easier for users to create matters and get useful insights
Problem: unreliable AI output
One major problem was that AI didn't always give the same answer, which could make it hard for prototype designs to work in real life. Another problem was finding the right balance between using contextual AI and handling auto-enabling correctly.
Design explorations
After exploring various design ideas, I reached a point where I felt confident enough. Ready to get feedback on the first designs.
First try: Prebaked prompts for welcome modal
Context-aware AI for collaborators
Context-aware AI for notifications
Prebuilt prompts for text
Getting feedback
During a feedback session, we concluded that the color I picked didn’t match the brand. It had to stand out, but the above purple was too different from Lupl’s teal identity.
A key takeaway from this meeting was to use familiarity. Since our customers were already using Microsoft Copilot, we decided on this card system for our assistant, as shown below.
Focus change
Although the first designs were inspired by Notion's context-aware AI assistant, things turned out to be different along the way. Due to a development focus switch, the new requirements for a minimum viable product had been much simplified.
The new needs included four types of prompts on the dashboard AI modal. These cards had buttons that let you quickly turn text into tasks in Lupl.
Improvement: Instead of displaying text prompts one by one, they were arranged in cards to make it easier for users to choose.
AI output was simplified to make development easier and create a basic version of the product to test.
Matter-related cards were simplified for the final iteration. They provide quick follow-up
actions for quickly creating items.
Task-related cards are simple and display only the essential information and a progress status.
Additional improvements
Unfortunately, in the end, this feature was completely deprioritized. But the iterative design process helped me gain valuable insights into meeting technical and business constraints.
It helped me develop an agile attitude and become more aware that not all design ideas will make it into production. And most importantly, it gave me an opportunity to learn and improve.
