Here’s your text with improved grammar and formatting in Markdown:

Research Recommendation

Inspiration

The need for efficient research recommendations to accelerate literature review and streamline the research process.

What It Does

This tool provides research recommendations by generating five relevant keywords based on user input. It then queries arXiv, retrieves related papers, and processes them using Retrieval-Augmented Generation (RAG) with Mistral Embed to find the most relevant recommendations.

How We Built It

We developed this system using:

• Backend: Django REST Framework
  
• Frontend: Vite + React + TypeScript + Tailwind CSS
  
• AI Components: Gemini (Google) for prompt generation and Mistral Embed for embedding-based retrieval
  

Challenges We Faced

• Optimizing request speed and handling timeouts efficiently.
  

Accomplishments We're Proud Of

• The system functions as expected, providing meaningful research recommendations.
  

What We Learned

• Combining RAG techniques with different AI models to improve retrieval and recommendations.
  

What's Next?

• Enhancing performance to make the recommendation process even faster.
  

Built With

• ai
• amazon-web-services
• amplify
• django
• gemini
• mistral
• python
• rag
• react
• typescript
• vite