Planit: Generative AI Urban Planner
Inspiration
The inspiration behind Planit came from the motivation of using Generative AI for a greater good. Urban planning is an essential and tedious process required in the development of a country's infrastructure and economy. Planit aims to ease and assist urban planners around the world by providing ideation and visualization into the possibilities of sustainable and economic urban planning.
What it does
Planit intelligently understands user inputs using Generative AI understanding to build an urban planning ideation plan for the provided location which can be anywhere in the world.
How we built it
Planit was built using JavaScript React and Python FastAPI stack. It uses Amazon Bedrock GenAI models and AgentCore Memory running on Amazon Elastic Container Service (ECS). The agentic logic was built using the Strands Agents SDK.
Challenges we ran into
The challenges that we run into while building Planit was the problem of providing an interactive map along with the urban planning recommendations given by Planit.
Accomplishments that we're proud of
We are proud of the fact that Planit is intelligently able to imagine urban development on live maps around the world, adding an ideation layer on top of an interactive map.
What we learned
We learned that it was possible to integrate a GenAI Agent to build ideation on live interactive maps.
What's next for Planit
In the roadmap for Planit, we want to work on improving the response speed, improving the UI design, adding image editing capabilities so a user can upload a map that they have and Planit can provide urban planning recommendations on top of that.
Built With
- agentcore
- amazon-bedrock
- amazon-web-services
- fastapi
- javascript
- python
- react
- strands-agents
Log in or sign up for Devpost to join the conversation.