AI Wattch is an open-source browser extension that tracks the energy use and carbon emissions of prompts on AI chatbots like ChatGPT and Claude. Powered by Antarctica’s One Token Model (OTM), it detects the model used, estimates real inference impact, and provides daily summaries and prompt-level tips to enable more carbon-aware AI usage.

A curated collection of research, tools, and best practices for environmentally sustainable AI development and deployment.

Sustainable Modular Furniture Inspection

Plant Growth Predictor is a web application that uses machine learning to estimate the growth stage of a plant based on environmental factors like soil type, sunlight, water frequency, temperature, and humidity. It helps users make informed decisions about plant care and agricultural planning. Built with React, FastAPI, and TensorFlow.

Predict electric vehicle (EV) growth over 3 years using real-world data and machine learning. The model learns adoption patterns from historical EV data. Results are shown in an interactive Streamlit dashboard. Includes county-wise comparison, forecasts, and clean visualizations.

Project HANSAL: Hybrid AI for Next-gen: Sustainable, Affordable, and Lightweight

A Python-based smart home automation system using YOLOv8 camera detection to reduce energy waste when rooms are unoccupied ( Without any intervention of Iot devices)