Personal AI Buddy Companion

AI-driven chatbot built with Gradio and Python to simulate a personalized digital companion. Designed for long-term engagement, with potential integration of deep learning and Langchain for enhanced personalization using user preferences, contextual memory from conversations, and assists with daily tasks, and reminders.

AI Nutritional Health Assistant: Personalized Guidance for Indian Diets

Built an AI-powered chatbot providing personalized nutrition recommendations based on users’ health metrics and dietary preferences, tailored for 3+ Indian regional cuisines.

Integrated LangChain with Gemma to support 100+ context-aware queries enabling seamless conversation.

Achieved 90%+ accuracy by implementing the RAG pipeline using FAISS on a structured nutritional dataset of 3000+ Indian dishes.

Personal Finance System for Expense Tracking and Analysis

Engineered a financial system handling 10,000+ transaction records, with visualization of spending patterns.

Applied Isolation Forest to detect spending anomalies with 80% precision improving financial awareness.

Designed a microservice architecture deployed via Docker & Kubernetes, achieving scalability with 99.9% uptime.

Administrator Controlled Timetable Based Automatic Facial Recognition Attendance System

Developed a timetable-linked facial recognition system using OpenCV and ESP32-CAM automating subject-wise attendance tracking for 50+ students.

Achieved 75% recognition accuracy under varying lighting conditions, reducing manual errors.

Secured Best Presentation Award at SYMPOSIUM-2023 among 50+ teams, recognized for innovative use of embedded AI.