• 

  Generates knowledge graphs

• 

  Indexes papers

My Research Project Story

Inspiration

During my scientific research, I realized that a significant amount of time is spent reading research papers, taking notes, organizing literature, and developing ideas. This workflow is fragmented and inefficient, which inspired me to build a system that integrates all these tasks and forms a personal knowledge base. The goal is to make research more structured, efficient, and insightful.

What I Learned

Through this project, I learned how to combine AI technologies with research workflows. I explored AI-powered indexing, knowledge graph generation, and contextual note-taking. I also deepened my understanding of how to structure research metadata such as publication venue, datasets, links, and methods, enabling more effective discovery and connection of insights across papers.

How I Built the Project

I designed a system that:

1. Indexes papers with structured metadata for quick retrieval.
2. Generates knowledge graphs to capture research lineage and topic similarity.
3. Supports contextual note-taking, linking papers, insights, and methods into a cohesive knowledge base.

The system leverages AI agents to automate discovery, connection, and note-taking. It is intended to allow researchers—even beginners—to quickly learn from and relate multiple papers, forming insights efficiently.

Challenges

One of the main challenges is enabling Gemini to access real-time, networked information, such as citation counts. Additionally, implementing a workflow that allows batch processing of multiple papers for fast learning, linking, and discovery is technically complex. Balancing real-time querying with efficiency and scalability remains an ongoing challenge.

Built With

• css
• typescript