-
The landing page of the website
Inspiration
After speaking with several Morgan and Morgan workers and other attorneys about the most painful part of their jobs, we set our sights on making the process of client intake easier. In particular, our initial goal was to make the intake process much more efficient. As it stands, most intake happens over the phone, while workers fill out forms for calling clients. To resolve this, we wanted to make a webapp or AI chatbot to make the process much smoother for both parties. After talking to more Morgan and Morgan workers, we also decided to also use AI to generate a variety of analytical metrics for use by the attorneys, including potential monetary gain, urgency rating, and likelihood of resolve in trial or settlement.
What it does
QuickCase automates the client intake process. Once the client data has been inputted it is analyzed by AI to determine estimated monetary gain, urgency of the case, and probability of going to trial vs. settling outside of court.
How we built it
QuickCase is built from LangChain, Flask, HTML/CSS, JavaScript, Python, and Bootstrap. We had a Flask backend server that ran the AI code (with LangChain) and it also rendered the frontend with mostly pure HTML and CSS.
Challenges we ran into
We originally wanted an AI chatbot feature but unfortunately it had to be cut due to time. Throughout hours of prompt engineering and changing agent types, toolkits, tools, and more we could not get it to do what we wanted.
Accomplishments that we're proud of
We are especially proud of our frontend. We used only pure HTML and CSS to stylize most of the web page. Even without using a framework like React we were still able to make it look great and accessible to users.
What we learned
We learned patience and communication. Organizing information in a way that everyone can understand is only beneficial. If we had taken some extra time to communicate among our teammates some issues could have been resolved sooner rather than later. We also learned the importance of a good night’s sleep.
What's next for QuickCase
We would like to integrate QuickCase with industry standard software like Litify. Litify in particular has a very robust API suite, and it would be entirely feasible to integrate our solution into the software Morgan and Morgan already uses. Additionally, we would like to finish the chatbot feature for user input, as we had intended to add that from the beginning.
Log in or sign up for Devpost to join the conversation.