Docket

Inspiration

Law students work with complex cases containing facts, legal issues, judgments, timelines, documents, and evidence. However, the process of understanding and studying this information is often fragmented across notes, PDFs, documents, and separate study applications.

We wanted to approach this problem differently.

Instead of building another generic AI chatbot, we wanted to create a complete intelligent workspace where AI becomes part of the student's actual legal-learning workflow.

This led to Docket — an AI-powered legal learning and case management platform designed specifically for law students.

The idea was simple: a student should be able to take a complex legal case, organize its information, understand it with AI, revise it interactively, and prepare for an exam without constantly moving between different tools.

What it does

Docket combines AI-powered legal study tools with structured case management in one web application.

Students can create and organize cases, store important case information, attach exhibits, build timelines, connect related cases, and track their study progress.

Using Google Gemini, Docket can transform case information into structured learning resources such as:

AI Study Briefs
Case Facts
Legal Issues
Case Holdings
Case Chronology
AI-generated Flashcards
Self-Test Questions
Revision Material

The platform also includes:

Case Management
Case Search
Interactive Case Timelines
Exhibits for images, videos, and PDFs
Case Tags and Case Constellation
Study Streaks
Exam Timer
Deadline Tracking
Revision Sheet Export

The resulting workflow connects the different stages of legal study:

Case Information → AI Analysis → Structured Understanding → Active Revision → Exam Preparation

The key idea is that AI is not treated as a standalone chatbot. It is integrated into the application's workflow and works alongside traditional productivity and learning tools.

How we built it

Docket was developed as a full-stack web application.

The frontend uses React and Vite, with a responsive interface designed around the workflow of law students.

The backend uses Node.js and Express.js, providing REST APIs for case management, authentication, study resources, timelines, exhibits, and other application functionality.

We use MongoDB with Mongoose for persistent application data.

For generative AI, Docket integrates the Google Gemini API. Case information is passed through structured AI workflows to generate study briefs, flashcards, self-testing material, and other educational resources.

For multimedia case exhibits, including images, videos, and PDFs, we use Cloudinary for cloud-based storage and media management.

Authentication uses JWT, while passwords are securely hashed using bcrypt. Protected API routes help ensure that users can only access their own case information.

We also implemented automated testing and GitHub Actions CI to improve reliability throughout development.

The application brings these components together into a single end-to-end system rather than treating AI, data management, and study tools as separate products.

Challenges we ran into

One of the biggest challenges was making generative AI useful inside a structured application.

A chatbot can generate a response, but Docket needed AI output that could become usable learning resources. This required designing prompts and application logic that consistently produced structured information suitable for study briefs, flashcards, and self-testing.

Another challenge was integrating multiple technologies into a reliable end-to-end system. Docket combines a React frontend, Node.js and Express backend, MongoDB, Gemini, authentication, cloud storage, and automated testing.

We also had to design around different types of legal information. Cases can contain structured details, long text, timelines, and multimedia exhibits. Making these different forms of information accessible without creating an overwhelming interface required careful UX decisions.

A further challenge was deciding where AI should and should not be used.

We learned that adding AI everywhere does not automatically improve a product. AI provides the most value when it reduces repetitive work or transforms information into something useful. Features such as timelines, tags, exhibits, dashboards, and exam tools provide the structure that allows the AI functionality to become part of a meaningful workflow.

Accomplishments that we're proud of

We are proud that Docket evolved into a complete working product rather than a simple AI demonstration.

The platform connects case management, generative AI, information organization, visualization, and study tools into one workflow.

One of the workflows we are particularly proud of is:

Case → AI Study Brief → Facts & Issues → Chronology → Flashcards → Self-Test → Revision

This demonstrates how a generative AI system can be embedded into a larger application instead of existing as an isolated conversational interface.

We are also proud of building supporting systems around the AI, including case timelines, exhibits, Case Constellation, study tracking, exam tools, authentication, cloud storage, and revision sheet generation.

From an engineering perspective, bringing the frontend, backend, database, AI service, authentication, storage, testing, and deployment together into one functional application was one of our biggest accomplishments.

Most importantly, Docket focuses on a clearly defined real-world problem and audience: helping law students manage and learn complex legal cases more effectively.

What we learned

Building Docket taught us that creating an AI-powered product requires much more than connecting a language model to an interface.

We learned how to integrate generative AI into an existing workflow, structure AI outputs for downstream features, design APIs between different application layers, and build a user experience around the strengths and limitations of AI.

Throughout the project, we gained practical experience with:

Full-stack application development
React and Vite
Node.js and Express.js
MongoDB and Mongoose
REST API architecture
Google Gemini API integration
JWT authentication
Cloudinary storage
Automated testing
GitHub Actions and CI
Responsive UI/UX design
Application deployment

We also learned an important product-development lesson: technical complexity does not automatically create user value.

The strongest solutions are often the ones where the technology disappears behind a simple and useful experience.

For Docket, this meant using AI where it provides a clear advantage while allowing traditional software features to handle organization, navigation, visualization, and study management.

What's next for Docket

Our next goal is to continue improving Docket as an intelligent legal learning and case management environment.

We want to improve the reliability and usefulness of AI-generated educational content while making the overall learning workflow faster and more intuitive.

We also want to learn from real student usage and feedback, using that information to improve the existing experience rather than adding unnecessary complexity.

In the longer term, Docket could evolve into a broader intelligent knowledge environment where students can build, organize, understand, and actively test their legal knowledge from one place.

The core philosophy will remain unchanged:

AI should make complex information easier to understand and act on — not make the user's workflow more complicated.

Docket — Turn complex legal cases into smarter study. ⚖️🧠