Project Info

JustDoeIt

Devpost

Inspiration

As Berkeley students, we noticed a recurring struggle — finding the perfect study spot. Whether it was Haas being full, Moffitt too loud, or cafés lacking outlets, the process of locating a good place to study often felt like a daily quest. We realized that despite Berkeley’s countless study spots, most students rotate between the same few. The idea for JustDoeIt came from a simple question: “What if your study habits could recommend your next favorite spot?” By combining data, community insights, and AI, we wanted to make studying feel easier, smarter, and more personalized — helping students save time and thrive academically.

What it does

JustDoeIt is your personal study companion — an intelligent platform that connects UC Berkeley students to their best study environments. It helps you: Discover study spaces based on AI-powered recommendations Track your study sessions, streaks, and productivity Visualize real-time busyness and space availability Explore an interactive campus map of Berkeley’s top spots Learn from community ratings and tips shared by peers In short, JustDoeIt helps students study more effectively — and enjoy the process along the way.

How we built it

We built JustDoeIt using a modern full-stack architecture: 🧠 Backend FastAPI (Python) for a lightweight, high-performance REST API Supabase (PostgreSQL) for authentication and real-time database sync Python 3.9+ for data processing and analytics logic AI recommendation engine leveraging user behavior data 💻 Frontend React 18 + TypeScript for a fast, responsive UI Vite for rapid builds and local development TailwindCSS + Radix UI for elegant, consistent design Interactive Map Layer using Mapbox (experimental) 🧩 Architecture Frontend and backend communicate via secure REST endpoints. Session data and productivity metrics are stored in Supabase, then analyzed for real-time insights and personalized study recommendations.

Challenges we ran into

Building JustDoeIt was an exciting but challenging journey. Some of the key hurdles we faced included: 🔌 Data Integration: Combining user session data with Supabase analytics in real-time without performance lag. 📊 Recommendation Algorithm: Designing an AI model using Claude API that adapts to user preferences dynamically rather than relying on static rules. 🧭 Map Interaction: Rendering an interactive map that balances usability with information density (outlets, WiFi, crowd level, etc.). 🧱 Frontend Scalability: Maintaining a fast and fluid UI as we integrated multiple complex components (charts, maps, analytics).

Accomplishments we're proud of

🎓 Built a working AI study-space recommender from scratch 🗺️ Designed an interactive Berkeley study map with real-time data 📈 Created a personal study analytics dashboard to visualize habits and productivity ⚡ Developed a full-stack system that runs efficiently on local and cloud setups ❤️ Received great feedback from Berkeley peers who tested the prototype

What we learned

Building JustDoeIt taught us invaluable lessons in both technical and user-centered design: How to balance data engineering with UX simplicity The importance of real-time synchronization for meaningful analytics How small details — like outlet availability or noise level — drastically influence user satisfaction How to design for scalability early on, preparing for Bay Area-wide expansion We also deepened our understanding of FastAPI, Supabase, and frontend performance optimization using Vite and TailwindCSS.

What's next

We’re just getting started. 🚀 Upcoming Goals 📱 Launch mobile apps (iOS + Android) for on-the-go recommendations 🌉 Expand to Stanford, UCSF, and other Bay Area campuses 🤝 Introduce study group and social matching features 🌙 Add offline mode and smarter habit-tracking AI Ultimately, our vision is to create a global platform that helps students anywhere find their best study environments — powered by data, driven by community, and guided by design. Study smarter. Explore more. JustDoeIt.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • AnthropicIn code
  • CSSIn code
  • ExpressIn code
  • FastAPIIn code
  • HTMLIn code
  • JavaScriptIn code
  • PythonIn code
  • ReactIn code
  • SQLIn code
  • SupabaseIn code
  • Tailwind CSSIn code
  • TypeScriptIn code

12 of 12 appear in the indexed code.

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

581 KB

Source files

93

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars