JobSheet started as a frustration. As a student juggling coursework and part-time work, I kept losing track of the hours I'd worked and, more importantly, what I was owed. I wanted something simple, fast, and mine.
The problem
JobSheet started as a frustration. As a student juggling coursework and part-time work, I kept losing track of the hours I'd worked and, more importantly, what I was owed. Spreadsheets felt clunky. Existing apps were either too complex or locked behind subscriptions. I wanted something simple, fast, and mine.
The research
Before writing a single line of code, I spent time looking at what was already out there. Most timesheet tools were built for teams with full payroll systems, manager dashboards, and approval workflows. Nothing really served the individual worker who just wanted to log a shift and know their total. That gap gave me confidence that the project was worth building.
Learning the stack
I'd been comfortable with HTML, CSS, and vanilla JavaScript, but JobSheet was my first real project in Next.js and TypeScript. I chose them deliberately. I wanted to push myself into frameworks I'd be expected to know professionally. The learning curve was real. Understanding the App Router, server vs. client components, and type-safe API routes all took time. I leaned on the official docs, a few Fireship videos, and a lot of trial and error.
For the database, I went with Supabase. It gave me a hosted PostgreSQL instance with a clean JavaScript SDK and built-in auth, perfect for a solo project where I didn't want to spend time configuring a backend from scratch.
What I built
JobSheet lets you log shifts with start and end times and an hourly rate. It calculates your earnings automatically and keeps a running total across all your entries. The UI is intentionally minimal: a form to add a shift, a list to review them, and a summary at the top. Authentication is handled by Supabase, so your data is tied to your account and persists across devices.
Where it stands
The core functionality works. You can sign up, log shifts, and see your earnings. It's live and deployed. There are rough edges — the UI could be more polished, error handling is basic, and there's no export feature yet. But it solves the problem I built it to solve, and it's the first project I've shipped that I genuinely use myself.
What I'd do differently
I underestimated how much time typing errors would cost me early on. In hindsight, I should have set up a stricter TypeScript config from the start, rather than loosening it to get things working and planning to come back to it later. I'd also write tests earlier, even basic ones, rather than relying on manual checks throughout. Those are the things I'll carry into the next project.
