Inspiration

We were frustrated by the limitations of traditional bookmarking tools they save links but not context. Often, we’d bookmark dozens of articles only to forget why we saved them in the first place. Second Brain was born from the idea of creating a more intelligent, question-friendly way to organize and retrieve web content. We wanted a tool that could act like a personal research assistant: capturing what we read, remembering it for us, and helping us find it again fast.

What it does

Second Brain is a browser extension that turns your bookmarks into an intelligent, searchable knowledge base. When you save a page, it captures the URL and a screenshot, sends that data to a backend, and stores the content in a structured way. Later, you can ask natural language questions like “What were the articles I saved about AI ethics?” and it returns relevant saved pages.

How we built it

We built SecondBrain as a Chrome extension + full-stack app in three parts:

  1. Extension
  • Captures page URL, metadata and screenshot on click
  • Sends payload to our Flask API
  1. Backend (Flask + Docker)
  • Scrapes text with Requests + BeautifulSoup (falls back to screenshot OCR)
  • Generates OpenAI embeddings
  • Stores metadata in PostgreSQL and vectors in FAISS
  1. Frontend (Next.js + Tailwind)
  • Displays bookmarks, tags and semantic-search results
  • Instant previews powered by vector search + keyword fallback

Why these choices? Flask/Docker for rapid iteration and easy self-hosting; PostgreSQL + FAISS for flexible queries + fast semantic search; Next.js for a snappy, SEO-friendly dashboard.

Accomplishments that we're proud of

  • Built a working prototype of an end-to-end pipeline: bookmark → extract → store → search → retrieve.

  • Implemented semantic search that returns relevant results even when the question doesn’t exactly match the saved text.

  • Designed a clean and intuitive UI that feels lightweight and functional inside the browser.

Built With

Share this project:

Updates