StationScope

first frontend page
chat page with interactive map

Inspiration

Finding apartments as an intern or new grad moving to New York City (NYC) is a daunting task. Four months ago when I came to the city for a new job, I was looking for a place to live. I would have many pages open on Zillow, late nights compiling spreadsheets, and found contradicting advice from all corners of the internet. I found ChatGPT alone not very helpful at suggesting me neighborhoods since I had little idea of NYC geography.

This experience, combined with feedback from other new movers to NYC, is why Daren and I created StationScope, an interactive chatbot robust against occasional hallucinations by pulling reliable data from the U.S. Census Bureau and MTA for quantitative data measures. It is a custom AI with weights that guide a user to their dream apartment near a subway stop, making a final recommendation with a dynamic report highlighting points from the chatbot conversation along with information grounded in real-world data.

Some assumptions:

NYC bus infrastructure does not substantively substitute subway mobility.
People want to live on a subway line that directly takes them to their workplace (or end destination), or involves only 1 small, non-time consuming transfer.
Using 2019-2023 ACS data requires an inflation metric to accurately assess 2026 and beyond rent price.

What it does

You can type in a desired address on the StationScope landing screen, and a geolocation API helps you find a valid address. The app identifies the closest station to your input address as well as additional close stations within 1/4 mile. Behind the scenes, the app pulls real estate price stats using a U.S. Census Bureau API for all stations on the same lines as the closest station. We use census tracts partially or fully within a 1/4 mile radius of each subway station and the American Community Survey (ACS). This information, once accounting for inflation, is fed into the bot as reliable quantitative data. You are prompted to click to on a chat button, after which you can see an interactive map to get a good sense of geographical layout and can converse with the custom AI agent. The bot will iteratively ask you questions and ultimately get you your desired neighborhood/stop and information about that neighborhood/stop in the form of a concise report. The ultimate goal is to educate the new mover to NYC about their options and make a structured recommendation, much like a consultant or real estate broker would. Anyone can understand the inputs and outputs to this app, even as the infrastructure is a web of Javascript, data science, and API links and endpoints.

How we built it

To develop the app, we had two main workstreams: the frontend + data science which Max handled, and the agentic bot via OpenRouter and map, which Daren handled.