CoreyEmilista - the one that is still a question
A Rails application built to explore a problem that has not been pinned down yet.
Emilista is 214 commits of working Rails application in search of a sharp enough problem. It is the most honest entry in this portfolio because it is the one that has not worked yet, and knowing when to stop pushing something forward is a decision Corey now has to make on real projects.
What this is, on the record
- Started
- 27/02/2026
- Commits
- 214
- Status
- Exploratory
- Who did what
- Built by Kristian Papadakis. Predates Corey. Currently paused rather than shipping.
- Built with
- Rails, Hotwire, Tailwind
The project
Emilista is the entry most portfolios would leave out.
It is 214 commits of working Rails application, built between February and May 2026. It is also paused, because the problem it solves has not been defined sharply enough to justify putting it in front of anyone.
What we did
Built it properly, which is what made the answer trustworthy.
There is a version of this where you sketch an idea, decide you are not sure, and shelve it having learned very little. Building something real generates a genuine test: does using this thing make the problem obvious? Here it did not, and that is a result rather than a failure.
What we decided
Not to launch it.
That sounds passive and it is not. There is a live, working application sitting there, and every instinct - sunk cost, the desire to have shipped something, the tidiness of a finished project - pushes towards putting it out anyway.
The counter-argument is simple. Launching converts a cheap, reversible open question into an expensive public commitment with users attached to it. If the problem statement is still vague, all launching does is add support obligations to a thing you have not figured out.
So it is parked, explicitly, with the status written down. Not abandoned, not quietly rotting in a folder, and not being nudged along for two hours a month forever.
Why this page exists
Because a portfolio where everything succeeded is not evidence, it is marketing.
Every operator has an Emilista. The question is whether it is visible and labelled, or invisible and still costing you attention. Making it the former is a small thing that says more about how the rest is run than another success story would.
What we decided
Stop, rather than launch something unclear
There is a working application here. The decision was not to push it out anyway. Launching a product whose problem statement is still fuzzy converts a cheap open question into an expensive public commitment, along with the users, support load and sunk pride that come with it.
Get the question right before the answer
The stage is problem definition, not build. That is uncomfortable to sit in when there is already a codebase, and it is the correct place to be. Better to hold an unanswered question than ship an answer that does not hold.
Do not let it quietly consume attention
The real risk of an unresolved project is not that it fails. It is that it keeps taking small amounts of time indefinitely while never being finished or stopped. Parking it explicitly costs less than leaving it ambiguous.
What got built
- A Rails and Hotwire application at 214 commits, built between February and May 2026.
- Enough working product to test the idea properly, which is what produced the honest answer.
What Corey runs now
Keeping it visible while parked
A paused project still appears in the portfolio view with its actual status, so it is a deliberate decision rather than something quietly forgotten.
No maintenance theatre
Nothing is scheduled against it. Recurring work on a project that is not moving is just a way of feeling productive about it.
Emilista taught the least comfortable lesson in the portfolio, which is that an operator has to be willing to say a thing is not working. It is the reason Corey reports honest status rather than optimistic status, and why "early" and "exploratory" are values it will actually use about your projects. An operating system that only ever tells you things are going well is not an operator. It is a cheerleader, and you can get those for free.
Run yours the same way
Everything above is one operator across a portfolio. Corey is that operator, packaged. 28 days on us, no card.