CiviScope

Inspiration

Government policy changes often feel distant and technical, yet they directly affect everyday finances. We were inspired to build CiviScope to make economic reforms understandable at a personal level, empowering citizens with clarity and data-driven insights.

What it does

CiviScope simulates how policy changes like GST revisions, fuel price hikes, or tax adjustments impact an individual’s income and expenses. It provides monthly and yearly financial impact, a vulnerability score, and a clear explanation of the results.

How we built it

We developed a FastAPI backend integrated with a regression-based ML model trained on synthetic financial data. A scenario simulation engine recalculates policy changes in real time, while SHAP explainability and an LLM generate personalized insights on the frontend dashboard.

Challenges we ran into

One major challenge was generating realistic synthetic financial data for training the model. Ensuring explainability while maintaining prediction accuracy and designing a neutral, non-political explanation layer also required careful engineering.

Accomplishments that we're proud of

We successfully combined predictive modeling, real-time simulation, and explainable AI into a working MVP. The system delivers transparent, personalized financial insights rather than generic policy summaries.

What we learned

We learned how to integrate machine learning with deterministic simulation logic effectively. We also gained experience in explainable AI and designing AI systems responsibly for civic-tech applications.

What's next for CiviScope

Next, we aim to integrate real public economic datasets, add region-specific multipliers, and expand support for small business simulations. Our goal is to evolve CiviScope into a scalable civic intelligence platform.

Built With

• fastapi
• numpy
• pandas
• python
• react
• scikit-learn
• shap