Skip to content
Selected work

Owned product · 2026 – Present

StudyAI

An AI study workspace that turns topics into structured notes, summaries, quizzes, and saved revision material.

View live demoopens in a new tabView public sourceopens in a new tab

01 Scope

The product context behind the work.

Problem

Students often jump between chat tools, documents, and scattered notes when they need to understand a topic, revise it, and test recall. The useful path is not just generating text; it is moving from a study prompt into notes, practice questions, and reusable study material without losing context.

My contribution

  • Owned product direction, the portfolio brief, acceptance criteria, and the review loop
  • Reviewed the notes -> practice -> review journey against the agreed product brief
  • Verified the public source includes Supabase authentication, Postgres-backed saved material, and per-user RLS policy definitions
  • Reviewed Gemini-powered notes, summaries, saved-library, activity, and quiz flows against acceptance criteria

02 Approach

Important engineering decisions.

StudyAI centers on a simple study loop: enter a topic, generate useful material, save what matters, and practice until the concept feels clear. Umar owned the product brief, acceptance criteria, and review loop.

The public repository verifies a JavaScript-based Next.js and React workspace using Supabase authentication and Postgres, with per-user RLS policy definitions for saved material. Gemini powers structured notes, summaries, and quiz output. Those are implementation facts visible in the source, not claims about broader tenancy, live deployment state, or operating scale.

The notes workflow keeps the prompt and output side by side, while the dashboard brings generated notes, saved material, AI activity, and quick actions into one reviewable workspace. Umar reviewed those flows against the agreed notes -> practice -> review direction.

The quiz flow completes the loop with difficulty and question-count controls plus an interactive multiple-choice session. The case study describes the delivered product and public proof while keeping Umar's role scoped to product ownership and review.

03 Evidence

Product screens from the shipped flow.

The notes generator turns a study prompt into structured revision notes with overview and key concept sections.
The quiz generator creates interactive multiple-choice practice from a topic, with difficulty and question-count controls.

04 Delivered & verified

What the public proof supports.

Delivered scope

Study flow

The product connects note generation, quiz practice, dashboard review, and saved study assets

Public implementation

The public source shows Next.js and React in JavaScript, Supabase Auth/Postgres with per-user RLS policy definitions, Gemini, and Tailwind CSS

Product-owner scope

Umar owned the brief, acceptance criteria, product review, and portfolio truth boundary

Stack

Next.jsReactJavaScriptSupabase Auth / Postgres (RLS)Google GeminiTailwind CSS

Reflections

  • The AI layer is most useful when the output is structured for studying, not just returned as a long answer.
  • A study product needs continuity: generate the material, save it, revisit it, and practice from it.

Next step

Have a similar project?

If this looks close to the product work you need, book a short call and I will help scope the next step clearly.

Book an intro callopens in a new tab