Inspiration
Local economic data such as footfall and business activity is often available, but extracting meaningful insights from it requires technical skills and dashboard expertise. We wanted to make this information accessible to everyone by allowing users to ask questions in plain English and instantly receive data-backed insights and recommendations.
What it does
Local Economic Pulse Dashboard is an AI-powered decision intelligence platform that enables city officials, local businesses, and citizens to interact with local economic data using natural language.
Users can ask questions in plain English, and the platform generates:
- A data-backed insight
- An actionable recommendation
- A transparent explanation of the reasoning behind the recommendation
How we built it
We built the application using Streamlit for the frontend and Google's Gemini API for natural language understanding and reasoning.
Workflow: Question → Gemini generates a database query → Query runs on the local footfall and business activity database → Gemini analyzes the results → Insight + Recommendation + Why.
During development, SQLite was used as the database. The architecture is designed to migrate directly to BigQuery and Cloud Run without changing the core application logic.
Challenges we ran into
One of the main challenges was translating natural language questions into accurate database queries while ensuring reliable AI-generated insights. Another challenge was Google Cloud billing verification during the hackathon, so we used SQLite for deployment while keeping the architecture compatible with BigQuery and Cloud Run.
Accomplishments that we're proud of
- Built a complete conversational AI interface for local economic data.
- Enabled natural language querying without requiring SQL knowledge.
- Generated explainable AI recommendations instead of only displaying charts.
- Designed the application to be easily scalable to Google Cloud services.
What we learned
This project strengthened our understanding of prompt engineering, AI-assisted query generation, explainable AI, and building end-to-end decision intelligence applications using Gemini API and Streamlit.
What's next for Local Economic Pulse Dashboard
- Integrate Google BigQuery for large-scale datasets.
- Deploy the application on Cloud Run.
- Connect real-time economic and mobility data sources.
- Improve anomaly detection and predictive analytics.
- Add support for multiple cities and real-time monitoring.
Log in or sign up for Devpost to join the conversation.