CityAnalysis

Inspiration

This project was inspired from the approach local governments are taking in the current age of data and artificial intelligence. As data scientist and business students, we have access to tools like IBISWorld, MarketLine (these are market research platforms), etc, which are fantastic, however, they only cover macro data, which misses the intricate micro data, that serve alot of value for businesses and organizations that seek expansion and opportunity in under covered markets.

CityAnalysis's main value proposition is to take the guess work out of micro-targeting, market expansion, and most important, give organizations the means to have a constant pulse on market conditions.

What it does

This platfrom does 3 main things:

• Runs a perpetual data infrastructure pipeline that extracts and reworks fragmented microeconomic data using our proprietary troybot.

  • The platform hosts the data in a structured and simplified manner in out cloud database.
  • The frontend is where users interact with the platform, querying from the database and make analytical decisions.

How we built it

We built it with a combinations of tools. For the frontend UI, we used nextjs. For the backend, we built autonomous bot with python that runs on a schedule to extract-transform-insert into our cloud database called supabase.

We utilized AI as well. Specifically to train and implement to "troybot" we selected and trained a claude agent for this task. We also used AI to generate UI components, but the logic flow of the platform was done manually because of issues that arose with data flow.

Challenges we ran into

The biggest change we ran to is the problem we are solving, data! Across the municipal portals, there is no standardization of data, hence we had a lot of trail and error building the bot, setting up the database to be consistent as possible, and displaying it interactively with a UI for a non-data centric user.

Also an initiale challenge was setting up the ETL pipeline and THEN figuring out how to automate it. This took manual work at first and a lot of iterations, but although not perfect, we got it to work

Accomplishments that we're proud of

• We are proud to have completed this one time and have a live version available.
  • we are proud to have the made an autonomous bot that searches the web and automates the ETL pipeline.
• Most of we are proud that we were able to work together to solve this pain point that we believe needs solving.

What we learned

• We learned to work as team. Specifically, planning our development project, flow and deciding who does what.
• We also learned that solving pain points like this takes alot of thought and alot of back and forth between what is currently available and what's possible.

What's next for CityAnalysis

• Expanding the reach of our automated system and improving it to work with more complex data.
• Furthermore, we want this to get in the hands o business professionals, mainly (to start), marketing firms that need micro data, developers, and investors that need an information edge to stay competitive.
• We also think that starting lines of comunication with municipalites and get contracted to get them up to date with current data extraction and data analysis technologies such as the one we have built.

Built With

• claude
• nextjs
• python
• supabase