I build AI systems that don't stop at the model.
- π Second-year B.Tech IT student at MIT Vishwaprayag University
- π Background in ENTC & Embedded Systems β I think about AI as infrastructure, not just models
- π‘ Currently studying Computer Networks β understanding how intelligent systems communicate and scale
- π€ Self-teaching AI, LLMs, RAG, MLOps & ML β entirely outside my college curriculum, driven by deep interest
- π Former Test Technician at Jabil β real hands-on hardware and electronics experience
𧬠Decision DNA
AI Monitoring & Explainability Dashboard for Credit Risk Models
- π Trained on ~5,000 loan applicants β Random Forest: 94% | Logistic Regression: 92%
- π Drift detection using Population Stability Index (PSI)
- π§ Explainable AI using SHAP
- π‘οΈ Adversarial threat classification β Low, Medium, High risk
- βοΈ Stack: Python Β· FastAPI Β· React Β· TypeScript Β· Scikit-learn Β· Gemini API Β· PostgreSQL
π VitaAI
AI-Powered Multilingual Healthcare Chatbot
- π Built on RAG pipeline for contextual accuracy
- π Multilingual support for diverse user groups
- π― ~90% accuracy in intent classification and contextual retrieval
- βοΈ Stack: Python Β· LLM APIs Β· RAG Architecture Β· FastAPI Β· NLP Β· Scikit-learn
AI & ML-Powered Civic Complaint Analysis and Management System
- ποΈ NLP-based complaint classification pipeline
- π Hotspot and priority zone detection using unsupervised clustering
- βοΈ Stack: Python Β· Scikit-learn Β· NLP Β· FastAPI Β· React Β· Pandas
IoT-Based Automated Irrigation & Remote Monitoring
- π§ Automated irrigation based on real-time soil moisture thresholds
- π‘ Remote monitoring via Firebase + Android App
- βοΈ Stack: Arduino Uno Β· ESP8266 Β· DHT11 Β· Firebase Β· Android
- π Generative AI and LLMs β IBM
- π Claude Code in Action β Anthropic
- π Machine Learning in Production β DeepLearning.AI
- π AI Infrastructure & Operations Fundamentals β NVIDIA
- π AWS Cloud Practitioner Essentials β Amazon AWS
- π Machine Learning Using Python β Infosys Springboard
- π Introduction to AI β Google
"Interest is what starts learning. Building is what proves it."