CaseFlow | Devpost

Agent pipeline
CaseFlow swipe interface
AI chat assistance agent
Mobile view

Inspiration

We saw an opportunity to bring modern interaction patterns and agentic AI into a traditionally manual decision-making process. We also particularly wanted to tackle this challenge because some of our team had previous backgrounds in AI agents and pipelining.

What it does

CaseFlow utilizes a swipe right/left interface to make the case decision process easier for lawyers. Our AI agent pipeline gathers case details and brings them all to the same place at a lawyer's fingertips. It also streamlines the case building and client contact stages.

How we built it

CaseFlow is a web application built with Next.js and Typescript using libraries like shadcn. We used Agentuity to create an agentic pipeline for research and analysis, as well as an AI chat agent. MongoDB supports our data layer, enabling case storage and retrieval.

Challenges we ran into

We faced challenges getting started with regards to MongoDB authentication and Agentuity organization and project structure. We also ran into issues with our projects building incorrectly despite code consistency among everyone. We managed to fix these through trial and error and some debugging.

Accomplishments that we're proud of

We are proud of our ability to plan our project out and execute our wireframe. We are also proud of our ability to work simultaneously on so many dynamic parts and connect them seamlessly. We all attempted a technical aspect that was new for us in some way!

What we learned

We learned more about Typescript layouts, UI frameworks, routing endpoints, database structure, agentic pipelines, and also law!

What's next for CaseFlow

CaseFlow has the potential to support law firms by making internal and external processes faster, smarter, and more engaging. Some of our future plans include user accounts, scheduling, automated client outreach, and deeper AI-powered case insights.