Focus forge AI

Inspiration

Students often know what they need to study, but struggle with how to begin. Static planners, templates, and Pomodoro timers don’t adapt to real behavior, making it easy to fall behind. We wanted to build something that feels like a personal study coach—one that structures the work, adapts to you, and keeps you focused without being intrusive.

What it does

FocusForge turns any study goal into a personalized, AI-generated workflow. It breaks topics into micro-tasks, creates time-boxed focus sprints, detects distraction patterns, and automatically adjusts pacing as you progress. Students get a smart, dynamic plan they can actually finish.

How we built it

We used Next.js and Shadcn UI for the front end, Firebase for authentication and storage, and OpenAI for task decomposition. A lightweight client-side engine tracks session data (tab switches, idle time, completion speed) and feeds an adaptive pacing module that recommends adjustments. Firestore stores workflows, tasks, sessions, and progress so plans update in real time.

Challenges we ran into

Designing a workflow schema that works for every subject.
Creating adaptive rules that feel helpful, not annoying.
Ensuring distraction detection is fully privacy-safe.
Keeping AI outputs consistent and validating them reliably.
Compressing the entire experience into an intuitive, minimal UI.

Accomplishments that we're proud of

Converting any study goal into a complete plan in under 3 seconds.
Building an adaptive engine that responds to real behavior.
A clean, student-friendly UI that works immediately without onboarding.
A demo that shows a complete end-to-end workflow from idea → plan → focused session → progress dashboard.

What we learned

We learned how powerful structured AI workflows can be when paired with real-time behavior signals. We also learned the importance of guardrails—validation, error handling, and fallback plans—to keep the system reliable even under unpredictable AI outputs.