ImpactAI: Project Story
About the project ImpactAI was inspired by the challenge of making decisions under uncertainty. We wanted a tool that could reason about scenarios dynamically, using real-world context, instead of static predictions or simple dashboards.
What I learned I learned how to:
- Use Gemini not just for text, but as a decision engine.
- Parse RSS feeds and incorporate live sentiment into scenario reasoning.
- Build interactive visualizations with graphs that reflect probabilities and risks.
How I built it
- Backend: Python + FastAPI, endpoints to fetch RSS feeds and call Gemini.
- AI: Gemini API generates outcomes, probabilities, reasoning chains, and sentiment-influenced adjustments.
- Frontend: HTML/CSS/JS with Chart.js to visualize branching outcomes and highlight risks.
Challenges faced
- Ensuring Gemini returned structured JSON reliably.
- Integrating RSS feed sentiment without slowing down the simulation.
- Designing an interactive graph that clearly communicates probabilities, outcomes, and reasoning.
In the end, ImpactAI lets users enter a scenario, pull in real-world context, and see interactive, AI-powered predictions — making decision-making smarter, faster, and more visual.
Log in or sign up for Devpost to join the conversation.