Machine Learning Engineer with a focus on building AI-powered solutions for environmental sustainability and social impact. Currently working on ecosystem restoration and climate technology projects.

Core Competencies:

• Environmental AI and climate technology solutions
• Machine learning system design and deployment
• Time-series forecasting and multi-objective optimization
• Geospatial analysis and environmental data science
• Open-source contribution to sustainability projects

AI-Powered Ecosystem Health Monitoring and Restoration Planning for Delhi, India

A comprehensive end-to-end machine learning platform that combines real-time environmental monitoring with AI-driven restoration scenario optimization. The system processes multi-source environmental data to provide actionable insights for ecosystem restoration planning.

Technical Achievements:

• Developed 5 production ML models with XGBoost achieving 99.75% accuracy (R² = 0.9975, RMSE = 2.21 µg/m³)
• Built interactive real-time dashboard using React and FastAPI architecture
• Processed and analyzed 18 datasets totaling 16,222 environmental records
• Engineered 89 features through advanced time-series analysis techniques
• Identified potential for 33% PM2.5 reduction through optimized intervention strategies
• Deployed production-ready REST API with 8 endpoints for real-time predictions

Technology Stack: Python, React, FastAPI, XGBoost, TensorFlow, Prophet, LSTM Networks, Tailwind CSS,

Data Sources: Based on IPCC AR6 framework utilizing NASA POWER API, World Bank data, and data from data.gov.in

Project Links:

• Repository
• Live Demo
• Technical Documentation

Model Performance:

• Achieved 99.75% prediction accuracy in ecosystem health modeling
• Developed production-grade XGBoost model with RMSE of 2.21 µg/m³

Data Engineering:

• Processed and analyzed 16,000+ environmental data records
• Integrated multiple data sources including NASA POWER API and World Bank datasets

Algorithm Development:

• Implemented 100+ optimized restoration scenarios using NSGA-II multi-objective optimization
• Developed custom feature engineering pipeline generating 89 predictive features

System Architecture:

• Designed and deployed full-stack ML platform with React frontend and FastAPI backend
• Built production-ready REST API handling real-time environmental predictions

Research Impact:

• Identified potential for 33% PM2.5 reduction through data-driven restoration planning
• Created framework based on IPCC AR6 climate assessment guidelines

Research Areas:

• Climate technology and environmental AI applications
• Geospatial machine learning for environmental monitoring
• Real-time data pipeline architecture for ecosystem health tracking
• Multi-objective optimization algorithms for sustainability planning

Professional Development:

• Advancing expertise in production ML system design
• Exploring edge computing for environmental sensor networks
• Contributing to open-source environmental and climate tech projects