Hello, beautiful people. I'm so excited to introduce ScholarAI, your AI research assistant.
First and foremost, a big thank you to Hashnode for hosting this hackathon.
Idea
As soon as I saw the Hashnode AI hackathon post, I started brainstorming ideas for an AI project. I explored different concepts, but none seemed interesting. Thanks to a recent conversation with my cousin, which sparked an idea. He mentioned he was having difficulty researching various topics for his research assignments. I thought, what if there was a web app to help him with his research? That's how I decided to build an AI web app that assists him and others with their research assignments.
Process
The moment I decided to build this project, the first step was to get access to the research papers available on the internet. I started searching for an API that could do this and was glad to found this one https://www.semanticscholar.org/product/api. This API provides access to more than 200 million papers.
The next step was to integrate an AI-powered assistant feature. I chose to use Langchain, which simplifies the development of AI-powered apps. For the embedding model and chat model, I decided to go with Prem AI and Groq.
For the note-taking feature, you need a good text editor. What could be better than TipTap?
ScholarAI Walkthrough
Try ScholarAI - https://scholarai.vercel.app/
Link to the repo - https://github.com/geekychakri/scholar-ai
Watch this short video to see how ScholarAI works.
Sign-up/Login using your GitHub account.
Search a research topic.
You'll be presented with a list of research papers based on your search. Click on any paper.
You can now view the PDF, create an AI assistant on top of any paper, and take notes as needed.
Bookmark any paper that you find helpful.
Tech stack
NextJS
Tailwind
Supabase
Langchain
Vercel
I had so much fun building this project. Hackathons like these really motivate you to build interesting and useful projects.
Thank you for reading the article, hope you all like the project. Please let me know your feedback in the comments section. Till then, be happy and stay safe.
Happy Coding 🚀