MeterMind started because, as college students always trying to save money, we kept asking “why is my electricity bill so high?” but we quickly realized this problem isn’t just about budgets: many people, renters or homeowners alike, lack visibility into which devices and habits drive their bills. MeterMind is an AI-powered, no-hardware web app that turns simple user inputs (home type, rooms, device usage) into a clear breakdown of monthly energy in kWh and cost, highlights the biggest energy “hogs,” and gives personalized, actionable recommendations. We built a working prototype with Python and Streamlit, modeled device energy using common power ratings and usage patterns, and layered in an AI coach (Gemini) to translate numbers into plain-language guidance. Along the way we wrestled with balancing model accuracy and simplicity, designing a friendly UX so users don’t need to know wattages, and making AI integration robust with fallbacks. We’re proud of shipping a live demo, making abstract bills actionable, and doing it without requiring smart meters or extra hardware. What we learned is that energy literacy is fixable with good framing and rapid prototyping, and next we’ll add real tariffs, optional smart meter imports, richer AI coaching, user accounts and tracking, and pilots with utilities, all while keeping the app approachable for students and everyone who wants clearer control over their energy.