Focus Tutor — Why We Built It
We didn’t start with a grand vision about reinventing education.
We started with a simple, slightly uncomfortable observation:
we all procrastinate in the exact same way.
You open your laptop to study.
You tell yourself you’ll “just check something quickly.”
Suddenly, 40 minutes are gone.
There are already thousands of AI tools that generate notes, summaries, flashcards, explanations. Content isn’t the bottleneck anymore.
Focus is.
That became the core idea behind Focus Tutor:
not another AI that explains things, but an AI that watches behavior in real time and protects your study session while it’s happening.
The Core Idea
We saw three problems:
- Tasks are too vague to start.
- Distraction happens unconsciously.
- Students don’t see their own focus patterns.
So we built something that lives directly in the browser and reacts to behavior in real time.
How We Built It
Focus Tutor is a Chrome extension (Manifest V3) with a small Node.js backend, using the MiniMax API for structured reasoning tasks.
Chrome Extension
→ Service Worker (tab tracking, timers)
→ Backend API
→ MiniMax API
It monitors tab switches and page context, nudges users when they drift, breaks vague goals into actionable subtasks, and generates short behavioral insights from real session data. All tracking stays local using chrome.storage.local.
What We Learned
- How to code as a real team — branching, merging, resolving conflicts, and building features in parallel without breaking each other’s work.
- How to connect to and structure API calls — designing prompts, parsing responses safely, and handling failures.
- How strict Manifest V3 is — service workers sleep, so persistence and state management require careful design.
The Point
Focus Tutor doesn’t try to make students smarter.
It tries to make it slightly harder to drift.
Instead of generating more content, it protects the time meant for learning.
Log in or sign up for Devpost to join the conversation.