Amicus is an AI-powered moot court simulator that transforms how law students prepare for appellate advocacy. Instead of static chatbots, it features an interactive "AI Judge" that actively cross-examines students on their legal arguments in real-time.
🛑 The Problem We Are Solving Law students and practitioners often lack realistic, high-pressure environments to practice oral arguments before stepping into a real courtroom. Practicing in front of a mirror or peers doesn't adequately prepare them for the sudden interruptions and rigorous cross-examination they will face from actual judges.
💡 What It Does (How It Works) Amicus acts as a virtual sparring partner. Users simply upload their moot proposition (case facts) and written memorial (legal strategy). Using Google's Gemini AI, the platform processes the case theory. When the user delivers their oral arguments into the simulator, the AI Judge actively listens for logical flaws, contradictions, or ignored facts, and ruthlessly interrupts them with piercing, skeptical questions—just like a real judge. Additionally, our "Rebuttal Assistant" acts as a co-counsel, predicting the opponent's strongest arguments and suggesting counter-strategies.
⚙️ How We Built It (Tech Stack) Frontend: Built with React 18 and Vite, styled using Tailwind CSS to create a custom, immersive courtroom-themed interface. Backend: Node.js and Express handle the server logic, utilizing multer for in-memory file handling and pdf-parse for document extraction. AI Engine: Powered by the Google Gemini SDK (gemini-3.6-flash) for rapid document comprehension and dynamic, real-time cross-examination generation. 🔮 What's Next for the Project In the future, we plan to implement voice-to-text integration so students can literally speak to the AI Judge instead of typing, add multiplayer capabilities for team practices, and include comprehensive scoring rubrics based on real-world advocacy standards.