LockedIn

Inspiration

Many students, including ourselves, struggle to stay focused during study sessions, often getting distracted by phones, fatigue, or wandering attention. While having study partners for accountability would be ideal, it’s not always realistic. We wanted to create a solution that brings that accountability into a personal, AI-powered platform—LockedIn—designed to help students maintain focus, track distractions, and build productive study habits.

What it does

LockedIn helps students stay focused and accountable during study sessions. Users can start a session via our website, logging in securely with Discord. Once a session begins, the system activates the user’s webcam regardless of the user's OS to track facial orientation and attention in real time. If a user looks away from the screen for extended periods, the platform immediately sends alerts to help them regain focus.

As an additional feature, users can link their phone number to receive AI powered custom text notifications whenever they are detected using their phone, further reducing distractions. LockedIn also provides real-time analytics, including total time spent distracted, phone usage duration, the number of distraction events, and the user’s current focus state, empowering students to monitor and improve their productivity over time.

How we built it

LockedIn combines AI, computer vision, and web technologies to help students stay focused. We used MediaPipe Face Mesh for head-pose tracking and YOLOv8 to detect phone usage in real time. The backend, built with Python and FastAPI, processes video asynchronously and triggers alerts via SMS APIs when distractions are detected.

On the frontend, we built a responsive platform using Next.js and TypeScript, displaying live analytics on focus, phone use, and distraction patterns. All session data is stored in databases, giving users actionable insights into their study habits. By integrating AI detection, real-time feedback, and analytics into a single platform, we created a seamless tool to hold students accountable and help them stay focused.

Challenges we ran into

During development, we encountered several technical challenges that tested our problem-solving skills. Early on, the application was connecting to the wrong camera—attempting to use a phone camera instead of the computer webcam. Resolving this required careful debugging and a deep dive into MediaPipe’s documentation to correct the syntax.

Another challenge was designing and managing the database to store user session data, analytics, and phone alerts efficiently. Ensuring real-time performance while handling multiple types of data pushed us to optimize our architecture and workflows.

Accomplishments that we're proud of

Over the weekend, our team built a platform that helps students stay accountable while studying—a project we’re especially proud of. We successfully implemented computer vision frameworks to track focus and phone usage, developed a responsive website with TypeScript, and integrated databases to manage session data in real time. Beyond the technical wins, we sharpened our problem-solving and critical thinking skills while strengthening teamwork and collaboration.

What we learned

Over the weekend, our team gained hands-on experience across multiple areas of full-stack development and AI integration. Our frontend teammates strengthened their skills building a responsive TypeScript website and integrating Gemini for real-time user interactions. On the backend and AI side, we learned how to connect computer vision models to live video feeds, process head and phone detection data, and even send automated text alerts to users when distractions were detected. Beyond the technical skills, we all improved our collaboration, problem-solving, and ability to quickly learn and implement new technologies in 36 hours.

What's next for LockedIn

We plan to continue developing LockedIn, refining the platform, and eventually releasing it to the public. Future improvements include launching our own domain, adding more AI-powered features to help students stay focused, and enhancing real-time analytics to provide better insights and accountability during study sessions.