Focus Buddy- Block distractions with AI

Inspiration

With online classes, remote work, and endless distractions, staying focused has become harder than ever. We noticed how often we lose track of time just by looking away from the screen or reaching for our phones. That sparked the idea: What if we had an AI friend who reminds us to focus the moment we drift? That’s how Focus Buddy was born a simple, smart productivity tool powered by real-time webcam monitoring.

What it does We used the following tools and technologies:

Teachable Machine (Google): To train a custom face detection model with two classes Focused and Distracted.
HTML/CSS/JavaScript: For building the user interface and integrating model prediction logic.
Webcam API: To stream live video input from the user’s camera.
JavaScript Logic: To track focused/distracted time and trigger alerts if the user gets distracted.

How we built it

How to train and export a custom ML model using Teachable Machine.
Using webcam input in real-time and connecting it to prediction-based logic.
Creating distraction timers and alerts using JavaScript.
Designing a user experience that feels encouraging, not invasive.

Challenges we ran into

Fine-tuning the ML model to work with different lighting and face angles.
Ensuring the distraction detection doesn’t throw false positives.
Making the alert system smooth and non-intrusive.

Outcome Focus Buddy successfully monitors user attention and delivers gentle nudges when focus is lost. It’s lightweight, requires no extra hardware, and can easily be used by students, remote workers, or anyone who wants a digital accountability partner.