How I Shipped 4 Apps in 4 Weeks
For years, I was the guy with half-finished projects.
A friend and I would meet after work, code for an hour or two, week after week, trying to build a startup together.
We kept it up for almost a year.
What did we have to show for it?
Zero products.
Just a graveyard of half-baked ideas and spaghetti code.
Life got in the way. Interest faded. Nothing shipped.
The Turning Point
Then the AI wave hit.
I bought my first desktop in a decade (GPU and all) and started hacking with local LLMs.
One weekend I wrote about using an LLM to hack a Japanese corporate exam.
It blew up on Reddit — comments, feedback, traffic spikes. I was hooked.
I realized two simple things:
- Sharing the story of what I build is as important as building itself.
- I can ship something in a day, and write about it the next.
So I gave myself a challenge:
Ship one app every week in September.
I called it #Shiptember.
Bootcamp & Big-Company Thinking
I came from the bootcamp world. I even taught there.
The goal was always the same: get hired.
Bootcamps (and most companies I worked for) train you to code like an employee: hyper-detailed, buzzword-heavy, obsessed with structure.
But building as an indie maker is the opposite:
Strip it down. Ship fast. Fix it live.
Users don’t care about architecture diagrams. They care if it solves their problem today.
My System: The Minimum Viable Loop (MVL)
To keep myself shipping, I built a system.
It’s a loop with three phases:
Ideation → Build → Show & Tell.
Here’s how it works:
1. Ideation
I jot down every idea that comes to mind - no judgment.
(Inspired by James Altucher’s “The Ultimate Guide for Becoming an Idea Machine”.)
Then I filter:
- Would I pay for this?
- Am I excited to build this?
- Can I explain it in one tweet?
If yes, it moves forward.
2. Tiny Core
Once I know the problem, I strip it down to the absolute minimum:
- Direct pain link: Does it solve the exact pain I green-lit?
- Use-today test: Would I (or a target user) actually use it today?
- 72h cap: Can it be shipped in three days or less (using my starter template)?
- One-sentence spec: "User does X → gets Y."" If I can’t say it simply, I kill it.
Everything else—polish, extras, integrations—comes later.
3. Show & Tell
-
Use the app every day and share one artifact/day (screenshot, short clip, before/after).
-
Aim for 7 posts in 7 days, tied to real usage, not mockups.
-
Recruit a tiny beta (10 people). DM them:
"I built X to solve Y. Want early access?" -
Give them one success path — a single clear flow:
entry → core action → outcome.
That’s it. No feature creep. No endless planning. Just proof it works.
Want the Full Playbook?
I’m writing a book called Ship Daily.
It’s in presale. I’m drafting chapters now, and early readers get them first.
If you want to help shape it as part of the early crew — and get the complete system behind #Shiptember: