Goodnotes LM
Streamlining 7× faster research speed for the Chevron Renewable Energy Group with AI
Role
Product Designer
UX Researcher
Category
AI
B2C
Tools
Figma
Notion
Collaborators
Product
Engineering
Timeline
3 months
March 2025 - May
Challenge
Goodnotes is a leading digital note taking platform used by millions of people worldwide. Students use Goodnotes to review their lecture notes, write practice problems, and build study guides.
However, students are increasingly turning towards new AI tools for their studying needs.
Mission
Build a proof-of-concept AI document analysis and outline generation platform. This platform would be used internally by Goodnotes to test the feasibility of implementing a similar tool.
Research
We conducted 6 user interviews and conducted a survey among college students to learn more about their study preferences.
We also conducted a competitive analysis of other note-taking apps to find our product-market fit.
User Flow
I outlined our user flow to understand our intended user journey and streamline our design process.
Ideation
With initial sketches and wireframes, I experimented with different ways of laying out the homepage and document analysis page.
Iterations
Our design went through many iterations, as we focused on forming consistent on-brand design components, improving accessibility, and finding non invasive ways to implement our features.
Design System
We designed a new, modern design system to reflect Goodnote's move towards AI tools. Our goal was to design visuals that were clean and simple while following AI design industry standards.
Final Designs
The landing page allows users to upload new PDFs to be analyzed and to view previous analyses. We used a card layout to organize users’ saved outlines. Within each card, the user can view the classification of their document and add custom secondary tags to personalize the way they organize their notes.
Turning segmentation on visually chunks out content from the document. The user is then able to ask questions about specific sections. This narrows down the scope and details of the document for the AI model, leading to more accurate responses.
The outline and summary generation sidebar showcases key sections from the uploaded document, allowing users to visualize main points from their document. Users can choose from multiple AI models to best support their study needs.
The chat feature allows users to ask questions about their document's content, or choose from preset prompts based on common use cases. With segmentation enabled, users can also ask questions about specific sections, allowing them to receive more focused answers.
With multi-model comparison, users can compare insights from different AI models side by side. This allows students to select AI models that best fit their current needs.
Takeaways
This project was very impactful in improving my visual design skills and knowledge of design systems. It also was a valuable experience in working with developers, product managers, and other designers.
Goodnotes LM
Streamlining 7× faster research speed for the Chevron Renewable Energy Group with AI
Role
Product Designer
UX Researcher
Category
AI
B2C
Tools
Figma
Notion
Collaborators
Product
Engineering
Timeline
3 months
March 2025 - May
Challenge
Goodnotes is a leading digital note taking platform used by millions of people worldwide. Students use Goodnotes to review their lecture notes, write practice problems, and build study guides.
However, students are increasingly turning towards new AI tools for their studying needs.
Mission
Build a proof-of-concept AI document analysis and outline generation platform. This platform would be used internally by Goodnotes to test the feasibility of implementing a similar tool.
Research
We conducted 6 user interviews and conducted a survey among college students to learn more about their study preferences.
We also conducted a competitive analysis of other note-taking apps to find our product-market fit.
User Flow
I outlined our user flow to understand our intended user journey and streamline our design process.
Ideation
With initial sketches and wireframes, I experimented with different ways of laying out the homepage and document analysis page.
Iterations
Our design went through many iterations, as we focused on forming consistent on-brand design components, improving accessibility, and finding non invasive ways to implement our features.
Design System
We designed a new, modern design system to reflect Goodnote's move towards AI tools. Our goal was to design visuals that were clean and simple while following AI design industry standards.
Final Designs
The landing page allows users to upload new PDFs to be analyzed and to view previous analyses. We used a card layout to organize users’ saved outlines. Within each card, the user can view the classification of their document and add custom secondary tags to personalize the way they organize their notes.
Turning segmentation on visually chunks out content from the document. The user is then able to ask questions about specific sections. This narrows down the scope and details of the document for the AI model, leading to more accurate responses.
The outline and summary generation sidebar showcases key sections from the uploaded document, allowing users to visualize main points from their document. Users can choose from multiple AI models to best support their study needs.
The chat feature allows users to ask questions about their document's content, or choose from preset prompts based on common use cases. With segmentation enabled, users can also ask questions about specific sections, allowing them to receive more focused answers.
With multi-model comparison, users can compare insights from different AI models side by side. This allows students to select AI models that best fit their current needs.
Takeaways
This project was very impactful in improving my visual design skills and knowledge of design systems. It also was a valuable experience in working with developers, product managers, and other designers.