My First Big Ask

On June 8, 2026, I started MyBidPilot.us with the kind of idea that sounds either ambitious or ridiculous, depending on how much coffee you’ve had.

The idea was BidPilot: an autonomous revenue platform for construction suppliers, steel building dealers, and subcontractors.

That is a mouthful. In plain English, I wanted software that could help a construction business take a quote request, understand it, turn it into a bill of materials, generate a proposal, follow up, and eventually create a project from it.

Basically: from “someone sent us a quote” to “we have a real project started.”

No small thing.

Especially for a 50-year-old non-coder sitting at a keyboard, asking an AI app builder to please make the machine do the thing.

The first big prompt

My first big ask was not modest.

I asked for the happy path version of the product. By “happy path,” I mean the simplest version where everything goes right. A customer uploads a quote. The AI extracts the important information. The app turns that into a bill of materials. Then it creates a proposal. Then that proposal can lead into a project.

I was not asking for every exception, edge case, user permission, pricing rule, or industry-specific rabbit hole yet.

At least that was the intention.

But even the “simple” version was big.

The construction world is not simple. Steel buildings are not simple. Subcontractor quotes are not simple. A quote can be messy, incomplete, full of assumptions, or written in a way only the person who created it fully understands.

So even as I typed the request, I had that little voice in my head saying, “This is too much.”

And honestly, it probably was.

What I asked it to do

At a high level, I asked the builder to create a pipeline.

Not a fancy oil-and-gas pipeline. A software pipeline. Meaning one step feeds the next step.

The flow was:

1. Upload a quote
2. Let AI pull out the key details
3. Organize those details into a bill of materials
4. Generate a proposal
5. Turn that into a project

I also chose to keep the first version on what I understood as the fastest path: TanStack Start with Lovable Cloud. I am not going to pretend I fully understood every technical implication of that choice. TanStack Start is basically part of the foundation the app sits on. Lovable Cloud is where the app can live and run without me having to become a server expert overnight.

That last part mattered a lot.

I was not trying to learn server management. I was trying to see if I could bring a construction revenue idea to life.

What came back

What came back was both exciting and sobering.

The AI did not laugh at me, which I appreciated.

It understood the general shape of the product. It gave me a starting structure for the quote upload, the extraction step, the bill of materials, the proposal, and the project flow.

That was the exciting part. There was suddenly something on the screen that resembled the idea in my head.

But it also made the size of the thing more real.

When an idea is only in your head, it can stay smooth and impressive. Once the app starts forming, all the hard parts show up. What exactly counts as a line item? What happens if the quote is vague? What information belongs in the proposal? What does “project” mean after the proposal is accepted?

The AI could build the bones. But I still had to make decisions.

That was the part I did not fully appreciate at the beginning.

Did it live up to expectations?

Yes and no.

Yes, because it helped me get from nothing to a working direction much faster than I could have on my own. Without the builder, I would still be staring at tutorials, wondering what half the words meant.

No, because it did not magically remove the work.

It did not take my big enterprise-sized dream and instantly turn it into a polished product. It gave me a first version of the path. A rough one. A useful one. But still rough.

I had to shrink my expectations from “autonomous construction revenue platform” to “can I get one clean quote to move through one clean process?”

That was humbling.

And probably necessary.

I almost wanted to keep adding more, because big ideas make you feel smart. But the smarter move was the smaller move: build the happy path first.

One quote. One extraction. One proposal. One project.

That became the first mountain.

Small takeaway: Big ideas are fine, but the first useful version has to be small enough to survive contact with reality.