MemoryVerse AI '26

🚀 MemoryVerse AI - Digital Identity System (Hackathon Submission) Overview: MemoryVerse is an AI-powered 3D Digital Identity Platform that ingests scattered resumes, certificates, project writeups, and internship letters, structuring them into an interactive WebGL Knowledge Graph, Timeline Trajectory, and Natural Language Vector Search Engine. ⚡ Quick Evaluation & Test Credentials: - Instant Demo Access: Click the "⚡ Instant Demo Workspace" button on the landing page or login screen to immediately explore pre-seeded evaluation data (14 connected documents across all categories). - Test Login Credentials: • Email: alex.rivera@identity.ai • Password: DemoPass2026!# • Role: Lead AI Engineer 🛠️ Local Setup & Running Instructions: 1. Prerequisites: Node.js (v18+), Python 3.10+ 2. Backend (FastAPI): cd backend pip install -r requirements.txt python main.py (Runs on http://localhost:8000) 3. Frontend (Angular + Three.js): cd frontend npm install npm start (Runs on http://localhost:4200) 🎨 Key Architecture & Design Decisions: - Interactive 3D WebGL Engine: Built with Three.js featuring 3D orbit navigation and 2D column flow visualization connecting Certifications ➔ Skills ➔ Projects ➔ Internships. - 768-D AI Vector Embeddings: Leverages Google Gemini embeddings and Supabase pgvector for sub-second natural language query retrieval and answer synthesis. - Auto-Classification Pipeline: Categorizes incoming PDFs, Word DOCX, and URLs into 6 identity pillars with automated skill and entity extraction. - Premium Modern Aesthetic: Designed using a dark glassmorphism palette, kinetic 3D tilt micro-animations, and HSL glow accents. 💻 Technologies & Tech Stack: - Frontend: Angular 22, Three.js, TypeScript, HSL Design Tokens - Backend: Python, FastAPI, Supabase (pgvector & Storage), Google Gemini AI API, PyMuPDF, Uvicorn