Lawgic

Inspiration

Spurred on by the Morgan and Morgan hackathon challenge we wanted to create a virtual assistant that can answer questions on a deeper level

What it does

Our bot is connected to a database of Mock Cases and will answer questions about these cases whether they range from case details, time ranges, or non-legal advice. As well as this the bot can also give you information on if you have a case or not based not on a case description

How we built it

We built it by fine-tuning a gpt-3.5 model for our specific use case for every user request.

Challenges we ran into

Database Management: Setting up and maintaining a robust database of mock legal cases proved to be a complex task. We needed to ensure data accuracy, consistency, and security. API Hosting: Hosting and managing the API for Lawgic presented challenges in terms of stability and scalability. We had to find solutions to handle a growing user base.

Accomplishments that we're proud of

We're proud of our ability to build our own API to store data, register users, and create a conversational model.

What we learned

We built upon our time management skills and greatly increased our capability to work as a team within GitHub

What's next for Lawgic

Higher Scalability and more users

Built With

• flask
• html
• react
• tailwindcss
• typescript