Project Info
This project did not submit a demo video on Devpost.
Inspiration
While learning about the housing process in Santa Cruz as a sophomore, I found it interesting that the best way to secure cheaper and reliable housing was through word-of-mouth, where maybe your friend's friend would mention that someone they know would be moving out soon. I soon realized that despite this being a primary mechanism of passing down housing, there wasn't any specific platform dedicated to making this process more efficient. In UC Santa Cruz, one of the biggest issues with being a student here is the prospect of housing, but maybe through a system like SlugShack, the notion of "for student, by student" housing can become much more prominent, securing housing for countless students.
What it does
Localhost website contains the user interface for the database and also the housing chatbot. This configuration allows a user to search through nearby available housing, and sort results by various metrics like distance to campus, # of vacancies, estimated monthly rent, and more. To the right of the intractable database is the chatbot, which can guide users to resources regarding affordable housing, explain some of the main causes of the SC housing crisis, and more.
How we built it
The database of available student houses is populated by a Google Forms Questionnaire, filled out by graduating/departing students who wish to indicate that there will be a future vacancy in their home. The form populates the Google Sheets database with information like the housing address, estimated rent, and other categories. The database processes this information with the Google Places API, automatically generating metrics like distance and driving duration from an address to the UCSC campus, and writes to the spreadsheet using the Gspread API. Additionally, we also developed a Santa Cruz housing chatbot AI with the Vectara Semantic Search platform and their APIs. We provided several housing studies, guides, and informational pages all pertaining to Santa Cruz housing to train the AI. We created a local host website with Streamlit to house the SlugShack code and the housing chatbot alongside each other. Used the Streamlit API to the database as an intractable display. To incorporate the chatbot into the Streamlit website, we used Flowise AI, which piped the Vectara-powered chatbot through a OpenAi wrapper, which finally enabled us to embed the chatbot in the html of the Streamlit website.
Challenges we ran into
For us 4, Cruzhacks is our first ever hackathon. All of us have mostly coded within our schoolwork assignments, meaning we hadn't been exposed to many of the methods and tools available in the every day tech world. For the project we were undertaking, it was extremely API heavy, and none of us had ever handled an API in depth. For us, this was our first proper project, first interaction with API's, first many things, and we struggled to find our footing. For one, we intended to use the Zillow API to query housing infomation automatically, only to find that it had been closed off recently. Because all of us are almost entirely Python programmers, we had no idea about web development, so we stayed away from it and settled on trying to make a user interface in a terminal. One of our biggest issues was with calling the Google Sheets API too often, leading to an automatic error that would halt all progress.
Accomplishments we're proud of
On that note, it forced us to learn how to maximize the efficiency of our program, such that it now has no issues with the Sheets Api. We were all super nervous about the project we had lined up for us, it seemed like an incredibly daunting task for a group of passionate but not very experienced coders. But in the end, when we settled for a terminal python interface, we managed to create a whole website!
What we learned
We learned an absurd amount about API's and their various forms. We learned about some of the systems that machine learning platforms use. We learned about databases through the various Google APIs, and we also learned how to set up a collaboratory Git setup.
What's next
specific market niche, similar level of connection as Indeed/famous hackathon projects Thinking about the idea of SlugShack, I realized that SlugShack has the potential to occupy a fairly niche part of the massive real estate market: student housing. There will always be students looking for housing, and students leaving their college homes. Just by changing a few variables and some other tweaks, SlugShack can be expanded to any other college town in America, which is an exciting possibility. It opens up a new real estate world: housing for students, by students.
JEVSCRUZHACKS2024
CRUZHACKS 2024 Project: SlugShack
Created in collaboration by: James Manlangit, Ethan Phan, Vishnu Naroth, Sofia Dang
Presentation: https://docs.google.com/presentation/d/1kROJYi_mQe_RrOY-w5lhHKKvklgCcrkuRApSUdg92yY/edit?usp=sharing
Description: SlugShack connects graduating/departing students in desperately needed houses with students seeking somewhere to call home. Chatbot coached on SC Housing Crisis studies explains complex housing system. A proof-of-concept of a tool that could alleviate the UCSC student housing crisis.
How it works: Website contains the user interface for the database and also the housing chatbot. This configuration allows a user to search through nearby available housing, and sort results by various metrics like distance to campus, # of vacancies, estimated monthly rent, and more. To the right of the intractable database is the chatbot, which can guide users to resources regarding affordable housing, explain some of the main causes of the SC housing crisis, and more. Note: All user information in the Google Sheets database is fictional.
Tools/APIs used: flowise github google-maps google-places gspread open-ai streamlit vectara
Credentials Needed:
Vectara API Key
Google Project API Key
Google: credentials.json
Vectara: token.json
Usage:
npx flowise start <- initializes localhost for Chatbot
streamlit run vectarainterface.py <- initializes localhost for Streamlit Website
Analysis
View
Metric
No commits on this project resolved to a GitHub account.
Technology
- PythonIn code
- OpenAIClaimed
- StreamlitClaimed
1 of 3 appear in the indexed code. 2 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.
AI coding agents
No AI coding agent signals were found in this repository.
Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.
Codebase size
Source size
13 KB
Source files
5
Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.
Repository
monkutten/JEVSCRUZHACKS2024
5 files · 13 KB · @ 6144947
Structure
Application logic
4 files · 80%Domain rules, services and shared utilities.
Supporting
Layers are inferred from where files sit in the tree, not from reading the code. A project that names its directories unconventionally will read oddly here — open the file browser to check anything the diagram implies.
Languages
- Python89%
- Markdown11%
Share of indexed source by file size. Binary and vendored files are excluded.
This project’s features have not been analysed yet.
Export this project's context (description, README, evidence, key source files) to chat with an AI agent elsewhere.