ME ESG Clean Energy Analytics Dashboard

Inspiration

The Middle East is the world's largest hydrocarbon exporter but also one of the regions most exposed to climate change. Saudi Vision 2030, UAE Net Zero 2050, and Qatar National Vision 2030 have committed hundreds of billions to clean energy. I wanted to actually track that with data.

What it does

Tracks ESG budget, renewable energy investment, CO2 reduction, and policy index across 8 countries and 5 sectors from 2015 to 2024. Includes Prophet-based forecasts through 2030 with confidence intervals, country deep dive pages, sector breakdowns, and anomaly detection.

How I built it

• Data pipeline in Python calibrated to real IRENA/IEA benchmarks
• FastAPI backend with 11 REST endpoints
• Prophet + Ridge Regression ensemble for forecasting
• React 18 frontend with Recharts and Bloomberg terminal aesthetic
• Separate Streamlit dashboard for exploratory analysis
• Deployed on Vercel (frontend), Render (API), and Streamlit Cloud

Challenges

Getting Prophet to run inside Docker on Render was tricky due to Stan compiler dependencies. Import path casing bugs in React (PascalCase directories on Windows vs Linux) caused silent failures on deployment.

What I learned

Deployment environments expose bugs that local development hides. Prophet is powerful but needs careful dependency management in containers.

What's next

Add real-time data feeds from IRENA and integrate green bond issuance data from Bloomberg Open API.

Built With

• docker
• fastapi
• plotly
• prophet
• python
• react
• recharts
• render
• scikit-learn
• streamlit
• tailwind-css
• vercel
• vite