IntellectAI

Inspiration

The inspiration for IntellectAI came from the need to make learning from static documents more interactive and engaging. We wanted to create a tool that not only allows students to study from PDFs but also enhances their understanding through interactive features like quizzes, summaries, and flowcharts.

What it does

IntellectAI enables students to upload PDFs and interact with the content in various ways. Users can ask questions, generate quizzes, create summaries, and even produce flowcharts based on the document's content. This turns a simple PDF into a dynamic learning experience.

How we built it

We built IntellectAI using a modern tech stack designed to maximize performance and scalability. On the frontend, we used Next.js to create a dynamic and responsive user interface. The backend was powered by Python with FastAPI, providing a robust and efficient API layer. For data storage and vector search, we utilized TiDB, which enabled us to manage large datasets and perform complex search queries efficiently. We have used JinaAI for embedding . Clerk was integrated for seamless authentication and user management, while Prisma was employed for database interaction and schema management. The AI-driven features were powered by Gemini API, enabling us to deliver intelligent recommendations and content processing. This combination of technologies allowed us to build a feature-rich platform capable of handling a wide range of user interactions.

Challenges we ran into

One of the main challenges was implementing the flowchart generation feature, as it required parsing complex information from PDFs and converting it into a visual format. Additionally, ensuring accurate and relevant quiz questions based on the document content posed a significant challenge.

Accomplishments that we're proud of

We are proud of successfully integrating multiple interactive features into a single platform. The ability to transform static PDF content into quizzes, summaries, and flowcharts marks a significant accomplishment. Another highlight is the smooth user experience, which allows students to focus on learning without technical distractions.

What we learned

During this project, we learned how to implement vector-search, enabling us to find and retrieve similar content efficiently. We also gained experience in creating embeddings using JinaAI, which played a crucial role in enhancing the accuracy of our search and recommendation features. Additionally, we explored various advanced techniques that contributed to the overall functionality and scalability of IntellectAI.

What's next for IntellectAI

Next, we are planning to enhance IntellectAI by enabling interaction with videos, allowing students to extract and engage with information in new ways. We also aim to add more advanced features to further enrich the learning experience

Built With

• clerk
• fast-api
• geminiapi
• jinaai
• langchain
• nextjs
• prisma
• python
• tidb
• typescript
• zod