SheSpace

Inspiration

Women often feel unsafe, isolated, or uncomfortable studying, walking, or existing in public spaces — especially on campus or at night.We wanted to build something that prioritizes safety, community, and comfort over just speed. SheSpace is not just about finding study spaces; it is about providing real-world safety signals, built by women, for women.

What it does

Safety Checks: Women can rate how safe they feel in a place based on specific criteria like lighting quality, staff presence, and inclusivity scores. The Support Score: These ratings generate a holistic "Support Score" that allows users to assess a location's safety in seconds. Gemini AI Insights: We utilize the Gemini API to analyze qualitative user reviews and generate "Safety Summaries," giving users a clear picture of the environment without reading dozens of comments. Walking Buddy: Users can also use the walking buddy tool to connect with verified peers heading in the same direction, ensuring no one has to walk alone.

How we built it

Frontend: We used JavaScript, HTML, and CSS to build the core application logic. We used React and Vite for a fast, responsive user interface, and Leaflet to render the custom interactive map layers.

AI Service: We built a lightweight Python (Flask & FastAPI) service specifically to handle requests to the Google Gemini API.

Challenges we ran into

One of our main challenges was integrating a React-based frontend with a Python FastAPI backend while incorporating an external AI service. Much of our time was spent troubleshooting environment setup, API configuration, and cross-service communication, including port management and request handling. We also encountered limitations related to AI model access and usage quotas, which introduced unexpected errors and required rapid iteration and adaptation. Working through these challenges pushed us to better understand backend–frontend integration and highlighted the importance of robust error handling and system design in real-world applications.

Accomplishments that we're proud of

Getting the custom color-coded pins (Green for Safe, Blue for Quiet) to render interactively was a huge win. Despite being beginners with API integrations and environment management, we successfully implemented Google Gemini AI, learning proper authentication, CORS configuration, and secure key storage practices along the way.

What we learned

We discovered that we don't need expensive enterprise tools to build professional maps. Leaflet gave us the flexibility to customize markers and layers exactly how we wanted.

We learned how to bridge a modern React frontend with a Python (Flask/FastAPI) microservice to effectively pass data to the Gemini API and get meaningful responses back.

What's next for SheSpace

Our top priority is moving from a simple login to a verified "Identity Tier" system (using tools like Onfido or Stripe Identity) to ensure every user on the map is a real, vetted individual. Additionally, expanding our Gemini integration to not just summarize safety, but to actively suggest the "greenest path" based on real-time community sentiment scores.

Built With

• fastapi
• javascript
• leaflet.js
• node.js
• react
• vite