Smart Energy Advisor

AI platform that analyzes energy consumption, suggests savings and shows climate impact, helping families and small businesses spend less and pollute less.

Comment 1

Inspiration

With climate change and rising energy costs, we wanted to create a tool that empowers people to make small but impactful changes in their daily energy use. The idea was to combine AI recommendations with an intuitive interface, so even users with minimal technical knowledge can save energy.

What it does

Smart Energy Advisor analyzes a user’s energy consumption profile and generates personalized, actionable tips to reduce waste and improve efficiency. The AI adapts suggestions based on usage levels — from very low to high consumption — to ensure relevance.

How we built it

We used:

  • React for the frontend (user-friendly dashboard).

  • Hugging Face Inference API with the Qwen model for natural language recommendations.

  • Vercel for deployment and secure API routing.

  • Markdown + icons + formatting for clear, engaging tips.

Challenges we ran into

  • Handling different response patterns from the AI and formatting them consistently.

  • Ensuring the API key is secure while still delivering real-time responses in a frontend app.

  • Time constraints of the hackathon forced us to simplify the MVP and prioritize the core features.

Accomplishments that we're proud of

  • Built a working AI-powered recommendation system in under 48 hours.

  • Designed a clean and simple UI that makes energy-saving approachable.

  • Learned how to integrate Hugging Face’s chat/completion APIs quickly and effectively.

What we learned

  • How to leverage AI models for sustainability use cases.

  • The importance of UX in presenting technical recommendations in a human-friendly way.

-How to balance innovation with feasibility under tight deadlines.

What's next for Smart Energy Advisor

  • Add data integration with smart meters and IoT devices for real-time analysis.

  • Expand to mobile apps for broader accessibility.

  • Use reinforcement learning to make tips more adaptive and personalized over time.

Built With

Share this project:

Updates