Pinpoint
Finding student housing in Kingston is brutal.
Listings are scattered across Facebook groups, Kijiji, and word of mouth. Roommates are found separately. Prices are often unclear until you click into posts. Location matters, but nothing is built around a map.
For Queen’s students and newcomers, housing search becomes chaotic, stressful, and time-consuming.
Pinpoint fixes this by bringing listings and roommates together on a single, shared map — powered by advanced AI.
What it does
Pinpoint is a map-first housing platform where listings and people looking for roommates live together in one view.
Instead of hopping between sites, users can:
- Explore rooms and units directly on a map
- See who is looking for roommates in each area
- Filter by budget, move-in date, lifestyle preferences, and more
- Describe exactly what they want, and let AI surface the most relevant options
- Add their own pin to indicate:
- “I’m looking for a place”
- “I’m looking for roommates”
This makes housing search faster, smarter, and far more intuitive, with everything in one place.
Why it matters for Kingston
Kingston’s student housing market is competitive and fragmented. Location and affordability are critical, yet current tools force students to manually piece together information.
Pinpoint changes that.
It creates a single, shared map where students can see housing options and potential roommates nearby. The result is reduced confusion, saved time, and an easier path to finding the right place — all powered by advanced AI to filter results intelligently.
How we built it
Pinpoint is a full-stack web application designed specifically for Kingston, with a focus on usability, data-driven search, and intelligent matching.
- Frontend: Next.js and React provide a fast, interactive interface centered around the map
- Mapping: The interactive map is the main interface — rooms, units, and people are all visualized by location
- Backend & Database: Supabase powers authentication, profiles, and data storage
- Authentication: Google OAuth ensures quick and secure login
- AI-Powered Search: Interprets natural language queries like “non-smoking 2-bedroom near campus under $1,000” and highlights the best results on the map instantly
- Real Data: Rental data is fetched from sources like Kijiji using polling and verified using multiple AI agents
Status
This isn’t just a demo — it’s a working MVP designed to realistically address the housing challenges people face in Kingston, while being scalable for other cities.
Built With
- google-gemini
- googleoauth
- leaflet.js
- nextjs
- postgresql
- react
- react-leaflet
- supabase
- tailwindcss
- typescript
- vercel

Log in or sign up for Devpost to join the conversation.