Home Safe

About us page
Landing page

Inspiration

Our inspiration was learning how 40% of adult Americans feel unsafe to walk at night near their home. We wanted to create a tool that could help Americans navigate local streets while also feeling safe. This tool will be used by navigation providers such as Google Maps and Apple Maps to provide safe but effective walking routes to their users.

What it does

The user uses natural language to explain their starting and ending destination. Our server processes this request and has an agentic workflow, where one agent finds ~10 routes and the other agent analyzes each for the recent crimes. The top three routes get displayed on our frontend with the recommended route taking up the embedded google map.

How we built it

We used Google's ADK and A2A to create agents that could query the maps api and also read and analyze data. Using sequential agents we create a system where one agent gets a lot of possible routes and the other analyzes and ranks them based on their safety. This all happens on a server while our front end sends requests for the user input and the output the agents give.

Challenges we ran into

We received Error 500 from the Gemini API a lot which slowed down development and limited our ability to test.

Accomplishments that we're proud of

We are proud that we implemented Google's ADK into a product that provides real benefit for American travelers by helping them choose safe walking routes.