Inspiration

School information is scattered across WhatsApp messages, emails, and notices. Important deadlines, events, and tasks can easily get lost in the noise. We built Scoop to turn these unorganized school messages into a clear, actionable plan for families.

What it does

Scoop uses AI to transform school messages into structured plan items, helping families keep track of important dates and responsibilities.

Capture: Paste a school message from a chat, email, or notice.
Extract: AI identifies relevant tasks, events, dates, and details.
Organize: Review extracted items, save them, and track completion.
Stay in control: Edit information before relying on it, with a fallback when AI extraction is unavailable.

Scoop focuses on one simple workflow: Paste → Extract → Plan.

How we built it

We started with Grok Build by xAI to explore the initial prototype, then developed and refined the product through our own implementation, debugging, and design decisions.

Scoop is built with React, TypeScript, and TanStack Start. Clerk handles authentication, Groq powers AI-based extraction, and PostgreSQL stores saved plan items. The application also includes a rule-based fallback for basic extraction when the AI service is unavailable.

We focused on keeping the experience straightforward while separating message processing, authentication, and saved planning data.

Challenges we ran into

Understanding unstructured messages: School notices rarely follow a consistent format. Extracting useful information from messages containing multiple events, incomplete dates, and ambiguous instructions required careful handling.
Making AI extraction reliable: AI can misinterpret dates or overlook important details. We worked to make extraction more practical, provide a fallback, and communicate failures rather than silently hiding them.
Building a complete workflow: Extracting information is only the beginning. We needed to connect authentication, data storage, and plan management so users could save items and return to them later.

Accomplishments that we are proud of

Built a working AI-powered school message organizer around a real everyday problem.
Connected natural-language extraction with structured, actionable plan items.
Implemented authentication and persistent storage for individual users.
Added a fallback extraction path for situations where AI is unavailable.
Focused on a usable end-to-end experience instead of an oversized feature list.

What we learned

Real-world text is unpredictable, and AI-generated results need validation.
A useful AI product needs more than a model; it needs a reliable workflow around it.
Authentication and persistent storage are essential when users depend on an application to organize their information.
Starting with one well-defined problem helps create a more focused product.

What's next for Scoop

Shared family plans so parents and guardians can coordinate school responsibilities.
Reminders and notifications for upcoming deadlines and events.
Improved date recognition and extraction accuracy.
Easier ways to bring school information into Scoop from existing communication channels.
Mobile-friendly capture for organizing school updates on the go.