I'm a Full-Stack Developer who ships production-ready, AI-integrated web applications — not just practice projects. My work spans from Next.js 15 apps with real users to IoT hardware running edge ML models at 98% accuracy. I bridge the gap between frontend polish, backend architecture, AI APIs, and physical devices.

"I don't just write code — I ship products that work, look great, and solve real problems."

constgulsum={role: "Full-Stack Developer | AI-Integrated Web Applications",location: "Virudhunagar, Tamil Nadu 🇮🇳",education: "MCA Graduate — Madurai Kamaraj University (2026)",liveApps: ["moonaichatbot.vercel.app","royal-mind-arena.vercel.app","heartcrafted-guls.vercel.app"],aiStack: ["Groq (Llama 3.3 70B)","Anthropic Claude API"],iotAchieve: "98% ML accuracy on physical ESP32-CAM hardware 🎯",led: "300+ students as MCA Department President 🎓",available: true// → hire me!};

Production AI chatbot with persistent multi-conversation history, 5 custom REST endpoints, markdown rendering, and animated glassmorphism UI.

• Architected 5 REST endpoints (POST /api/chat, GET/DELETE/PATCH /api/history/:id) for full conversation lifecycle management
• Integrated Groq's Llama 3.3 70B model with real-time streaming, persistent PostgreSQL storage via Prisma + NeonDB
• Built animated glassmorphism UI with full markdown rendering and responsive conversation switching

AI chess platform with Selen — a conversational AI opponent, full legal move validation, Drizzle ORM, and persistent game history.

• Built "Selen" — a conversational AI chess opponent powered by Anthropic Claude with full natural language chess commentary
• Implemented complete chess rule engine with legal move validation and persistent game history via Drizzle ORM + NeonDB
• Demonstrates adaptability across ORM tooling (Drizzle vs Prisma) in a live production environment

Smart IoT fence achieving 98% ML animal detection accuracy on physical ESP32-CAM. Edge Impulse model → Buzzer + Blynk real-time alerts.

• Final-year MCA project examined by faculty viva voce — trained and deployed edge ML model directly on physical ESP32-CAM hardware
• Completed hardware implementation at Hashan Medicare's technical training facility (Nov 2025 – Jan 2026)
• Built complete pipeline: camera capture → Edge Impulse inference → Blynk cloud alerts → buzzer activation

Desktop AI app with live webcam facial recognition, CSV export, duplicate-entry prevention, and persistent SQLite records.

• Integrated DeepFace for facial matching + OpenCV for real-time webcam video capture
• SQLite3 for persistent attendance records with duplicate-entry prevention and one-click CSV export
• Built full Tkinter GUI — no external UI framework, pure Python desktop experience

Luxury emotional gifting platform — memory boxes, love scrapbooks, and personalized keepsakes with a full ordering system.

Full Stack Dev & Cloud Intern — Futurik Technology, Madurai (Jul – Aug 2025)

• Built and deployed a complete Login Portal System on Netlify: responsive UI, JavaScript form validation, plan comparison table — end-to-end production deployment
• Gained hands-on cloud deployment workflows supporting the "Cloud Intern" component of the role

Web Development Intern(Vue.js Frontend)(Jun – Jul 2023)

• Completed training in Vue.js covering component-based architecture and reactive data binding
• Built mobile-responsive practice UI components applying mobile-first design principles

• 🥇 MCA Department President — Represented & coordinated 300+ students (2026)
• 🏛️ Union Member (2026) · Class Representative, 3rd Semester
• 🤖 Fundamentals of Machine Learning and AI — Coursera
• 💬 Prompt Engineering for ChatGPT — Great Learning
• 🐍 Python for Machine Learning — Great Learning
• 📊 Data Visualization using Python — Great Learning
• 🎨 Diploma & Certification in Animation Technology

I'm actively looking for Full-Stack Developer roles where I can build AI-powered, scalable products from day one.

📍 Sattur, Virudhunagar, Tamil Nadu · 📞 9361061756 · Available to join immediately