The latest articles on DEV Community by jaysid97 (@jaysid97). https://dev.to/jaysid97 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3645845%2Fd9725601-8733-4d5b-9092-a49e068f18c8.jpeg https://dev.to/jaysid97 en jaysid97 Mon, 20 Apr 2026 13:21:08 +0000 https://dev.to/jaysid97/ecotwin-an-ai-climate-coach-for-real-world-emissions-cuts-4dk7 https://dev.to/jaysid97/ecotwin-an-ai-climate-coach-for-real-world-emissions-cuts-4dk7 <p>What I Built<br> EcoTwin is a personalized climate action coach that turns a few everyday inputs, like commute habits, food choices, home energy use, and travel frequency, into a practical action plan with estimated annual CO2e savings.</p> <p>The goal was to build something useful immediately. Instead of giving users a generic sustainability lecture, EcoTwin acts like a climate twin: it estimates a baseline footprint, then recommends the highest-impact changes first.</p> <p>Users get:</p> <ul> <li>A baseline annual footprint estimate</li> <li>A personalized set of top climate actions</li> <li>Projected annual CO2e savings</li> <li>A concise AI coaching summary powered by Gemini (with fallback if key is unavailable)</li> </ul> <p>Demo<br> Live demo: <a href="https://ready-lamps-poke.loca.lt" rel="noopener noreferrer">https://ready-lamps-poke.loca.lt</a></p> <p>Quick walkthrough:</p> <ol> <li>Open the app</li> <li>Enter a city and lifestyle inputs</li> <li>Click Generate My Climate Plan</li> <li>Review before/after footprint and recommended actions</li> <li>Read the AI coach summary</li> </ol> <p>Code<br> Source code is in the project folder:</p> <ul> <li>Backend: Flask API for scoring, recommendations, and AI summary</li> <li>Frontend: HTML/CSS/JS dashboard with interactive results</li> <li>Data: Curated action library with estimated CO2e savings</li> </ul> <p>Core implementation highlights:</p> <ol> <li>Transparent footprint estimation model</li> <li>Rule-based personalization of actions by user profile</li> <li>Gemini API integration for natural-language coaching</li> <li>Stable local fallback summary for reliability</li> </ol> <p>How I Built It<br> I built EcoTwin as a focused full-stack Flask app to keep the product easy to run, easy to demo, and easy to judge.</p> <p>Architecture flow:</p> <ol> <li>Frontend collects a lightweight user profile</li> <li>Backend computes annual baseline emissions</li> <li>Action library is filtered by relevance and sorted by impact</li> <li>App calculates projected reductions and renders before/after metrics</li> <li>Gemini generates a short personalized coaching summary</li> </ol> <p>Design decisions:</p> <ul> <li>Kept input friction low so users can get value in seconds</li> <li>Prioritized practical behavior-change suggestions over abstract climate theory</li> <li>Added clear before/after visuals to make impact tangible during a live demo</li> </ul> <p>Prize Categories<br> This submission is for:</p> <ul> <li>Best use of Google Gemini</li> <li>Best use of GitHub Copilot</li> </ul> <p>Team<br> Solo submission.</p> devchallenge weekendchallenge ai earthday