★ Your Parallel Project — Build Whatever You Want
The homework project you choose, design, and build entirely on your own across Weeks 3–10. Any domain, any idea — as long as it's yours, it's real, and it ships by Demo Day.
Every week in class you build the School ERP together with the group. Every week at home you build your own thing — a project you chose, in a domain you care about, with no one telling you what it should do.
This is the most important habit this training builds: the ability to build something from zero, alone, with Claude Code as your only pair.
The One Rule
Ship it by Demo Day (Week 10 Friday). That's the only constraint.
What you build, what it does, how complex it is — entirely up to you. A working app that solves a real problem for a real person (even if that person is just you) is the goal. Polish is not required. Completeness is not required. Running, real, and deployed — that's the bar.
What Makes a Good Parallel Project
The best parallel projects share four qualities:
1. You are a user of it. You understand the problem without research. You know what "done" looks like because you live with the problem every day. CA professionals have a lifetime of problems that no software solves well — or solves them with clunky software they hate. That is your advantage.
2. The scope is small enough for homework. You have 2–3 hours per week. A project with 2–3 database tables, one or two personas, and a clear core loop is the right size. "An ERP for my CA firm" is not. "A tool to track which clients have filed their GSTR-3B this month" is.
3. It uses the same stack as class. React, TypeScript, Supabase, Next.js, React Query, Tailwind. You are reinforcing the class patterns — not learning a new stack from scratch on your own.
4. It would actually get used. Not a tutorial project you delete after completing. If you build it, add real data in Week 9, and it helps you in Week 10 — that is the right project.
What to Avoid
| Avoid | Why |
|---|---|
| A copy of the class project | You will not learn anything new — you already built it |
| A domain you do not understand | You will spend all your time learning the domain, not building |
| Too many integrations (Razorpay, email, SMS) from day one | Add those later — build the core first |
| A vague idea ("a platform for CAs") | No clear scope = you never finish |
| A project for someone else to maintain | If you are the user, you stay motivated |
Ideas to Spark Your Thinking
These are not suggestions — they are starting points to trigger your own idea. The best project is one that none of these describe exactly.
CA-domain ideas:
- A compliance deadline tracker for your clients (GSTR-1, GSTR-3B, ITR, audits)
- A simple GST invoice generator for a specific client of yours
- A payroll calculator that generates salary slips with TDS for a small team
- A document checklist tracker — which documents you've received from each audit client
- A meeting notes app — record what was discussed with each client, next steps, follow-up dates
- A fee register — what each client owes you, what's been paid, what's outstanding
Personal/general ideas:
- A personal expense tracker that categorises spending and shows a monthly summary
- A recipe app where you save your own recipes and plan weekly meals
- A habit tracker with a streak counter
- A simple inventory system for a family business (a shop, a workshop, anything)
- A reading list tracker — books read, rating, notes
The CA-domain ideas have one advantage: you understand the problem deeply, and you are the user. But if you have a personal project idea you've always wanted to build — this is the time.
How to Present Your Idea
Before you scaffold anything (by the end of Week 2), write three sentences:
- What it does: "A tool to track which clients have filed their monthly GST returns and flag who is overdue."
- Who uses it: "Just me — I'm the only user."
- What the main data is: "Clients (with GSTIN) and filing records (month, status, filing date)."
If you can write those three sentences cleanly, the project is well-scoped. Bring those three sentences to the Week 3 Monday session and Claude Code will scaffold the whole thing in one prompt.
The Three Support Pages
| Page | What it covers |
|---|---|
| Choose Your Topic | A step-by-step framework for finalising your project idea before Week 3 |
| Week-by-Week Guide | Generic weekly milestones that work for any project — with the CA compliance tracker as the worked example throughout |
| Bootstrap Prompts | The prompts to scaffold, set up Supabase, and start each feature — generic templates + a complete CA compliance tracker example you can follow directly |
If you genuinely cannot decide what to build, the CA compliance tracker (tracking GSTR-1, GSTR-3B, ITR deadlines per client) is a ready-made idea with a full worked example in this section. Use it — choosing quickly and building is better than spending two weeks deciding.