Shift

AI-personalized sustainability actions with radical transparency about AI's environmental cost.

The Problem

65% of people want to live more sustainably — only 26% follow through. The gap isn't motivation. It's tools.

Existing apps show guilt dashboards and generic tip lists. They don't tell you what to do today, in your city, with your diet, on your commute. And every AI product quietly ignores its own environmental footprint, making the problem worse.

How Shift Works

Shift onboards users in 90 seconds, then delivers one AI-personalized sustainability micro-action per day — tailored to:

• Your actual commute distance
• Your diet pattern
• Your city's live electricity grid carbon intensity
• Current weather conditions

Actions are grounded in EPA and DEFRA emissions data, structured using behavioral science frameworks (Fogg's B=MAP, Tiny Habits), and scored against 190 curated actions in a knowledge base. Users earn points, build streaks, advance through five levels, and track contributions to UN Sustainable Development Goals.

Radical Transparency

Unlike every other AI product, Shift shows you what the AI costs:

• Every action card displays the inference carbon cost alongside savings enabled
• A Chrome extension monitors the environmental impact of every Gemini prompt in real time
• A dedicated Eco-LLM dashboard tracks energy (Wh), carbon (gCO₂), and water (mL) per query
• Semantic caching serves similar queries without extra inference — zero additional carbon

Typical carbon ROI: 10,000:1 or higher.

Technology

• Frontend: Next.js 14 (App Router, PWA) · TypeScript · Tailwind CSS · shadcn/ui · Framer Motion · Tremor
• AI: Groq (Llama 3.3-70B) with Gemini fallback via Vercel AI SDK
• Database: Supabase (Postgres + pgvector)
• Caching: Upstash Redis (TTL cache) · Upstash Vector (semantic deduplication)
• APIs: Climatiq (commute CO₂) · Electricity Maps (live grid intensity) · Google Maps Distance Matrix · Open-Meteo
• Carbon Estimation: EcoLogits model with Groq LPU efficiency multiplier
• Analytics: PostHog · Sentry
• Email: Resend

Potential Impact

At 100,000 daily users completing one action each, Shift removes an estimated 12,000 tonnes of CO₂ per year — equivalent to taking 2,600 cars off the road.

The Eco-LLM transparency layer is also a standalone product for enterprise teams navigating AI carbon disclosure requirements under EU AI Act regulations.

Getting Started

# Install dependencies
npm install

# Set up environment variables
cp .env.example .env.local
# Fill in your API keys

# Run development server
npm run dev

Post-Hackathon Updates

Updates made after the initial hackathon submission:

• 2026-04-17: Fixed actionFrequency type mismatch (numeric → categorical), added missing API response fields, improved dashboard data fetching, added error toasts to onboarding, removed unnecessary retry logic.
• 2026-05-21: Finalized adding two new features to the extension: 1) Prompt Compression, and 2) Smart Image Handling (image compression for uploads and web search alternatives for image generation requests)

License

MIT

Background

Shift v0.1.0 was engineered in 12 hours for the GDG @ Penn State Solution Challenge to solve a critical irony: using AI to save the planet while ignoring its energy cost. It balances personalized climate action with full LLM-usage transparency.

Team: Suryansh Sijwali, Nabeel Ahmed, and Neil Barbara.