Priyadharshini

Full-Stack Developer

Chennai Institute of Technologyinternship

Open to roles

CommunicationLeadership

MemoryVerse AI '26

MemoryVerse AI ’26 is an AI-powered Digital Identity Platform that transforms scattered academic and professional documents into a structured, searchable, and connected digital knowledge repository. Unlike a traditional cloud storage system, MemoryVerse AI understands a student’s journey by analyzing uploaded resumes, certificates, project reports, internship documents, achievements, and other academic or professional records. Key capabilities: • AI-powered document text extraction and intelligent categorization • Automatic summary generation, skill detection, keyword extraction, entity recognition, and confidence analysis • Semantic search using embeddings and vector similarity • Interactive knowledge graph connecting skills, certifications, projects, internships, and achievements • Chronological digital journey timeline • AI Career Assistant that provides personalized strengths, career-readiness insights, skill-gap guidance, and recommended next steps • Original uploaded documents remain accessible through the platform Technology stack: • Frontend: Next.js 15, TypeScript, Tailwind CSS, Framer Motion, React Flow, Recharts • Backend: Node.js, Express.js, TypeScript, JWT authentication • AI Service: Python, FastAPI, Gemini API, NLP, embeddings, and RAG-based retrieval • Database: PostgreSQL with pgvector support • Storage: Local development storage with an architecture designed for Supabase Storage integration Setup instructions are available in the repository README. For local testing, start the Backend API on port 5000, the FastAPI AI service on port 8000, and the Next.js frontend on port 3000. The application can use a development fallback when PostgreSQL is unavailable; however, PostgreSQL with pgvector is recommended for persistent storage and full semantic-search functionality. Suggested reviewer flow: 1. Register a new student account. 2. Upload a resume, certificate, or project document. 3. Open the document analysis page to view the AI-generated summary, category, skills, keywords, entities, insights, and confidence information. 4. Use semantic search with a query such as “Show my AI projects.” 5. Explore the Knowledge Graph and Digital Journey Timeline. 6. Open the AI Career Assistant and ask questions such as “What are my strongest skills?” or “What should I learn next?” The central experience of MemoryVerse AI is: “I never have to search through folders again because my AI already understands my journey.”

5 media files · github.comView

QueueCure Pro – Real‑Time Clinic Queue Transparency

Clinic queues are confusing and stressful. Patients lack visibility into their position, causing repeated questions and frustration. Receptionists struggle to manage multiple registrations, while doctors face delays due to unclear queue status. Average consultation delays reached 20–30 minutes. A transparent, real‑time system was needed to reduce uncertainty and improve efficiency. Process I froze scope early to avoid feature creep. Built with React + Tailwind for responsive UI, Node.js + Express for APIs, MongoDB for persistence, and Socket.IO for real‑time updates. Focused on three public pages and three role dashboards. Iterated from 6 taps to 2 for patient tracking. Switched from static refresh to live socket events for instant queue changes. Results 1)6→2 taps to goal: Patients track their token faster. 2)91% task success rate: Usability testing showed patients could reliably track queue status. 3)70% reduction in patient uncertainty: Feedback showed patients felt more confident about wait times. 4)Receptionist efficiency improved: Token generation and queue actions streamlined into one dashboard. Reflection Next time, I would expand testing with actual clinic staff and patients instead of mock users. I would optimize the mobile experience further for patients tracking on phones. Admin charts were basic; I would add predictive insights such as peak hour forecasting. Finally, I would improve accessibility with multilingual support and larger display options for elderly patients.

11 media files · figma.comView

This is Priyadharshini’s work on Wooble.

Build a profile that shows what you can do — and share it anywhere.

Build yours