I have three projects sitting in a very awkward stage right now.

They work enough that I can see what they could become.

But they are unfinished enough that I can also see everything wrong with them.

Samantha is my attempt at building a social media management system around the way I actually want to create content.

Happy is a health and weight-loss app I am building for myself — something that looks beyond just calories and considers food, weight, habits, sleep, energy and my menstrual cycle.

And FIN is my attempt to finally see my entire financial life in one place — income, spending, savings, investments, CPF, debt, assets, liabilities and the goals I'm working towards.

None of them are finished.

Some buttons don't work.

Some features mostly exist because I told an AI:

"Eventually, I want this to do this."

Some parts look surprisingly polished.

Other parts very clearly belong to the category of:

we'll deal with that later.

And yet I've decided I want to start putting them on the internet.

That makes me deeply uncomfortable.

Because showing the idea is easy

I love ideas.

Ideas are safe.

When something exists only in my head, it can be incredible.

Samantha can become a system that helps me manage multiple social accounts, figure out what to post, organise content, understand what's performing and eventually turn content into income.

Happy can become the health companion I wish existed — one that understands that hunger, sleep, energy, exercise, nutrition, weight and your menstrual cycle don't exist independently of each other.

FIN can become the place where I open one screen and finally understand:

Where is all my money?

How much am I actually worth? How much debt do I have left? How much did I spend this month? Am I investing enough? How far am I from my goals?

In my head, all three work beautifully. The interfaces are beautiful. The integrations work. The recommendations are intelligent. Nobody clicks a button and discovers that absolutely nothing happens.

Ideas are nice that way. They haven't encountered reality yet.

The moment you put something in front of another person, however, it becomes real.

Someone can misunderstand it. Someone can think it is boring. Someone can point out an obvious flaw. Someone can ask:

"Why wouldn't I just use an existing app?"

Which is a completely reasonable question and also possibly the most terrifying sentence you can say to someone who has spent three nights building something.

My instinct is to wait until they're impressive

I keep catching myself thinking:

I'll share Samantha once the analytics actually work. I'll share Happy once the tracking is smarter. I'll share FIN once all my financial data connects properly. I'll post about them once they look more professional. I'll show people once I can confidently explain exactly what each one is.

Basically:

I'll show everyone once there is nothing embarrassing left to show.

The problem is that I'm starting to suspect that moment doesn't exist.

Every time I improve something, I notice five more things that could be improved. The better the project gets, the clearer its potential becomes. And the clearer its potential becomes, the more inadequate the current version feels.

If I follow that logic indefinitely, I could spend a year quietly polishing projects nobody knows exist.

Which would be comfortable. But probably not very useful.

I'm learning to separate "unfinished" from "not worth showing"

This distinction has been surprisingly difficult for me.

When I see an incomplete feature, I tend to interpret the entire project as incomplete. But those aren't actually the same thing.

A product can be unfinished and still contain an interesting idea. A prototype can be messy and still demonstrate something valuable. A project can be nowhere near commercially ready and still be worth talking about.

And sometimes showing the rough version is how you discover what the finished version should even be.

I'm already experiencing that with Samantha. I originally thought:

I'm building a social media management app.

Then I started building it. And I quickly discovered something important.

A beautiful dashboard filled with buttons is not a product. The buttons actually have to do something.

Obvious? Extremely. Still an important lesson when AI can generate an interface so quickly that, for a moment, your brain goes:

Wow. I made software.

Then you click "Schedule Post."

Nothing happens.

Right. Software.

Building with AI creates a strange illusion of progress

This is probably one of the most interesting things I've learned so far.

AI makes it astonishingly easy to make something look finished.

You describe what you want. A few minutes later there is a sidebar. Charts. Cards. Navigation. Beautiful gradients. Buttons. Suddenly you are staring at something that looks suspiciously like a SaaS product.

And that visual progress is incredibly motivating. But it can also trick you.

Because there is a huge difference between:

"I built a screen that represents this idea."

and

"I built a system that reliably does this."

I'm somewhere between those two stages with all three projects. And I think that stage is actually worth documenting.

Because I'm not really trying to prove that I can build apps

At least, not yet.

I'm trying to learn how products become products.

What should Samantha actually do that makes someone's life easier? Which parts of Happy would I genuinely use every day? What information inside FIN actually changes my financial behaviour, rather than just giving me prettier charts?

Those questions are becoming much more interesting to me than:

Can I make another feature?

Because with AI, the answer to that question is increasingly: probably.

The harder question is:

Should I?

These projects started because I wanted things that worked differently

None of these ideas started with:

"I want to launch a startup."

They started with irritation.

I wanted to create content consistently, but existing systems didn't quite match the way I wanted to work. So Samantha appeared.

I wanted to lose weight and understand my health without reducing every decision to a calorie number. So Happy appeared.

I wanted to see my income, investments, spending, CPF, debt, savings and financial goals without assembling the picture manually every time. So FIN appeared.

And somewhere along the way, I realised I had started building software around my own frustrations.

Maybe some of those frustrations are uniquely mine. Maybe they aren't. I don't know yet.

That's partly why I want to start sharing them.

There is also a slightly more selfish reason

I think putting my work on the internet will force me to finish things differently.

When nobody can see a project, it is incredibly easy to keep expanding it. One more feature. One more dashboard. One more integration. One more clever idea. You can build forever.

But the moment I have to explain Samantha to another person, I suddenly need to know what Samantha actually is.

When I show Happy, I need to explain why it exists. When I show FIN, I need to demonstrate what problem it solves.

Publishing creates a constraint. It forces me to turn:

"Look at all the things this could eventually do."

into:

"Here is what it does today."

I think I need that.

And yes, part of this is about becoming comfortable being seen trying

This might actually be the hardest part.

There is a very particular kind of safety in only showing people things you're already good at. Finished work. Polished presentations. Successful outcomes. Things where you already know the story ends well.

Building publicly removes some of that protection.

There is a chance Samantha never becomes a real product. Happy might change completely. FIN might turn out to be useful only to me.

Six months from now I may look back at these versions and wonder why on earth I thought they were good enough to post.

But maybe that's exactly why I should post them.

Because I don't really want a record that makes it look like I magically knew how to build everything. I'd rather have a record of figuring it out.

The ugly versions. The wrong decisions. The unnecessarily complicated solutions. The breakthrough moments. The:

"WAIT. IT WORKS."

followed approximately seven minutes later by:

"Never mind."

That feels much more interesting.

So I'm going to start showing the unfinished versions

Not because they're ready. They're not.

Not because I suddenly became immune to embarrassment. Definitely not.

But because these three projects represent something I want to get better at:

Taking an idea out of my head. Making something real. Putting it in front of people. Learning from what happens. And then making it better.

Samantha, Happy and FIN are currently three very unfinished experiments.

Maybe one becomes a product. Maybe all three do. Maybe none of them do.

But I don't think I want to wait until I know the ending before I start telling the story.

So this is me publishing them while they're still messy.

Broken buttons and all.