Juri Flow

Inspiration

My friend's brother is going through a serious legal case, and I really wanted to help him. When I learned that GPT-4 not only passed the bar exam but also ranked in the 80-90th percentile, I recognized a huge opportunity. Fortunately, I had access to the GPT-4 API, so I began brainstorming and learning about vector databases, embeddings, and Langchain on YouTube by watching James Briggs's videos.

What it does

Juri Flow - Your Personal AI Lawyer

• Legal Documents (chat with it — ask questions, gain valuable insights, locate key information etc)
• Consult with 10+ expert lawyers across diverse specialties powered by GPT-4

How we built it

Juri Flow is built using Langchain, OpenAI's embedding, Pinecone, and NextJS with Tailwind for the frontend.

Challenges we ran into

We didn't know anything about Langchain, Pinecone or embeddings, so I spent weeks learning about it.

Accomplishments that we're proud of

When we uploaded the document pdf, we were finally able to chat with it! Our friend's family was so impressed that they paid $150 to use the web app.

What we learned

We live in an era where we can leverage these tools to profoundly transform people's lives. We learned so much about this tech: Langchain, Pinecone, and OpenAI's embeddings.

The big goal is to provide top-notch legal assistance at a fraction of the cost, making exceptional legal support accessible to everyone, everywhere.

What's next for Juri Flow

• Building a mobile app for iOS and Android.
• Create a WhatsApp/ Telegram chatbot that uses OpenAI calling functions and Piencone and Langchain

Built With

• amazon-web-services
• langchain
• nextjs
• openai
• pinecone
• vercel