Predicting Carbon Intensity from Grid Frequency Data Using FFT Analysis and Gradient Boosted Regression

An advanced machine learning system that leverages ensemble methods and deep learning for comprehensive climate risk assessment and prediction. The tool combines multiple ML models with real-time weather data analysis to provide actionable insights for climate-related threats.

Hackathon winner at AI Engineer World Fair Hackathon: Transforming code, one function at a time, to reduce digital carbon footprints and create a more sustainable digital world.

PHREEQC geochemical modeling of Enhanced Rock Weathering in Darjeeling tea plantations for carbon removal quantification and agricultural soil enhancement

Decentralized environmental monitoring network on Cardano – VerdantWatch Catalyst Fund15 proposal

ML-based estimation of datacenter power demand and grid impact analysis for US electrical infrastructure.

Google Open-Source Project: Stochastic building simulator and real-world dataset for training and benchmarking reinforcement learning agents in energy-efficient smart control environments. Built with Gym, TensorFlow Agents, and 6+ years of real building data.

🌞 AI-powered detection of solar panels in satellite imagery using YOLOv8 with 97.3% accuracy

A curated list of high-quality resources, startups, tools, and frameworks driving innovation in climate technology.

Smart agriculture backend API that predicts optimal crops using Machine Learning. Features: Real-time weather integration, multi-language support, PDF reports, and historical yield analysis. Built with FastAPI, scikit-learn, and deployed on Render.

Open-source 16.7 kWh phase-change thermal energy storage system for building electrification

Dockerized FastAPI service for forecasting Germany’s hourly electricity load using real Open Power System Data (OPSD). Includes time-series feature engineering, machine-learning baseline model, and production-ready API endpoints.

AI-powered wildfire damage detection using IBM/NASA Prithivi-EO-2.0 foundation models and hybrid spectral analysis for high-precision burn mapping.

EnerCast is a data-driven machine learning project that estimates national CO₂ emissions from energy consumption patterns. It combines a reproducible data pipeline, leakage-safe feature selection, and rule-based baselines to compare traditional estimation approaches with ML models and evaluate emission-reduction scenarios for carbon credit scoring.

🌱 Real-time carbon footprint tracker for Python code. Measure your code's environmental impact and optimize for sustainability.

NDVI-based wildfire risk analysis using Sentinel-2 imagery, OSM powerlines, and geospatial ML

🌿 Monitor environmental health with decentralized sensors that provide real-time, tamper-proof data on air, soil, and water quality using Cardano's blockchain.