Markly Writing

Inspiration

Markly Writing helps students build structured, repeatable writing habits through rubric-driven feedback and short practice cycles. The product is currently student-first, focused on helping learners improve through targeted practice and clear feedback.

What it does

Guides students through scaffolded writing tasks with explicit structure prompts.
Evaluates submissions against teacher-authored rubrics and returns targeted, actionable feedback instead of opaque scores.
Supports a student-first writing workflow centered on practice, feedback, and improvement.

How we built it

Frontend: Next.js with React App Router and server-side rendering for fast lesson and dashboard pages.
Backend: NestJS API with Prisma for Postgres, background workers for scoring and PDF generation, and Firebase for auth where needed.
Integrations: rubric-driven scoring, PDF/report generation, email notifications, lightweight analytics, and migration/seed scripts.
Dev workflow: pnpm monorepo with TypeScript end to end and utility scripts for local development, migrations, and seeding.

Challenges we ran into

Modeling rubrics so teachers could author flexible criteria without breaking scoring logic.
Mixing Firebase on the server and client required lazy initialization to avoid SSR credential leaks.
Scaling background scoring and report generation while keeping the teacher experience responsive.
Migrating real classroom data safely and building reliable seeding tools.

Accomplishments that we're proud of

A student-first writing experience that focuses on clear, actionable improvement.
Reduced teacher grading time through automated, rubric-aligned feedback.
End-to-end integration across scoring, progress tracking, and classroom exports.
Practical dev tooling for migrations, seeding, and local development.

What we learned

Teachers respond better to rubric-language feedback and actionable next steps than raw scores.
Lazy auth initialization is essential when mixing SSR and client bundles.
Early investment in migration tooling prevents fragile manual fixes later.
Small automation, like batch scoring and seed data, delivers outsized productivity gains.

What's next for Markly Writing

Add adaptive learning paths that generate follow-up exercises based on rubric weakness.
Improve scoring transparency with explainable feedback snippets and exemplar comparisons.
Expand the student experience with more guided practice and clearer next-step feedback.
Polish the core workflow with faster iteration loops and better writing support.

Built With

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.