Learnspace

• 

  hero

• 

  about

• 

  features

• 

  q1

• 

  q2

• 

  q3

• 

  q4

• 

  q5

• 

  q6

• 

  setup guide

• 

  chrome extension also lets you manually add links, along with automatically storing new bookmarks added to the learnspace folder

• 

  learnspace populated with clusters

• 

  cluster selected

• 

  content generation example for text learner

• 

  content generation example for audio learner

• 

  search by topics

• 

  flashcard q1

• 

  flashcard a1

• 

  edit your links

• 

  developer dashboard for testing generation

Inspiration

One of our team members is super lazy and adds bookmarks on interesting/useful content all the time and then never revisits them again. We thought that was a waste of potential learning opportunity.

What it does

Learnspace is a Chrome extension AND web application that turns bookmarks into a learning system. Save links to your Learnspace folder in your bookmarks bar, and the app uses AI to extract insights, group related content into clusters, and generate personalized study materials. It supports text summaries (bullet points, paragraphs, or deep dives) and AI-generated podcasts, tailored to your reading style and background (are you a professional or a college student?). The dashboard visualizes clusters as interactive modules you can search, drag, and explore.

How we built it

We built a Next.js app as well as a Chrome extension that syncs bookmarks via the chrome.bookmarks API. The backend uses Snowflake for data storage and Gemini for information extraction, semantic clustering, and content generation. The clustering service groups related links by topic, and the content generation API creates personalized materials based on user preferences from an onboarding survey. The dashboard uses a mind-map design with custom animations and a search system that highlights and focuses matching clusters.

Challenges we ran into

Implementing incremental clustering that preserves existing clusters while adding new bookmarks required careful state management. Personalizing content generation to match user preferences (text format, jargon level, podcast style) was also a challenge. Building a smooth, performant visualization with drag-and-drop, search, and focus modes required optimizing renders and CSS animations. Integrating the Chrome extension with the web app while handling bookmark sync and API calls reliably was also a challenge.

Accomplishments that we're proud of

We built a full-stack learning platform with AI-powered clustering and personalized content generation. The visualization dashboard provides an intuitive way to explore and interact with learning clusters. The system handles incremental updates without losing user-generated content, and supports both text and audio learning styles. We integrated Gemini for intermediate representation (IR) extraction, clustering, and content generation, into a cohesive workflow.

What we learned

We learned how to use Gemini’s multimodal capabilities for semantic clustering and content generation. Building a Chrome extension that syncs with a web app required careful handling of permissions and API design. Working with Snowflake for data persistence taught us about managing structured data for AI-powered projects. The project reinforced the importance of user experience in making AI features accessible and useful.

What's next for Learnspace

We want to add more content formats like video summaries and interactive quizzes. Advanced analytics could track learning progress and suggest new topics based on interests/existing and past bookmarks.

Built With

• chrome.bookmarks
• gemini
• next.js
• snowflake
• wondercraft