← All articles
Projects

NewTechForge: From idea to production, and why those are two different jobs

Having an idea is easy. Turning it into a product that runs in production is a craft. That is why we founded NewTechForge.

The NewTechForge website has a single sentence on it: Forging new technology products from idea to production. The sentence is short, and I worked on it for a long time.

Two companies, two jobs

devpoint builds AI solutions for companies. A customer comes with a problem, and we solve it in their system. NewTechForge is something else. That is where our own products come from. An idea that we or a partner have becomes something you can eventually sell, operate and develop further.

That sounds similar, but it is a different job. In a customer project there is someone who knows what they want. With a product you have to find that out yourself, and you almost never find it the first time.

The idea is the cheapest part

I get to hear a lot of ideas. Good ones, and honestly also many I have heard before. That is not the problem. Fact is: ideas are cheap, and since AI they have become even cheaper, because in an hour you have a prototype that looks like it works.

What is expensive is the path from that prototype to something that still runs at three in the morning when nobody is watching. That can handle real data, which is never as clean as in the demo. That someone can still understand and change in two years.

A prototype proves that something works. A product proves that it lasts. Almost all the work lies in between.

Why forging is the right word

Forge. I chose the word because it describes how the process feels. You have a piece of material, and you work it until it has the shape it needs. Not until it is beautiful. Until it holds.

For software that means: build, put it in production, see what breaks, adjust, again. Whoever plans a product to completion on the drawing board and then builds it gets a very nice drawing board.

How a product comes about with us

The question: Which decision should become faster or better for whom? If we cannot say that in one sentence, we do not build.

The smallest core: Not the product, but the one thing that makes it valuable. We build that first, and in a way that runs in production, not in a demo.

Real users, early: Better five people who really use it than fifty who like it.

Forging: What breaks gets made stronger. What nobody uses gets thrown out. This is the part that takes longest and shines least.

Operations: A product is only finished when it runs without us. Monitoring, costs, retraining, support. That is part of it, not something that comes after.

What NeoCoder has to do with it

NeoCoder itself came about this way. From a problem we had: fifty micro models and several projects that could no longer be kept up by hand. A tool we built for ourselves, and which was then forged until others could use it.

By now it is the tool we use to build products at NewTechForge. Architect agent, coding agents, QA, artifacts. And it is the tool I use to bring teams into the world of AI Driven Development. Both are the same thing: from idea to production, with as few detours as possible.

To close

NewTechForge is the place where ideas become products. Not because we have the better ideas. But because we know how long the path from idea to production is, and walk it anyway.

How this text was made

Written by me. The thoughts, the values, the learnings, the mistakes: all mine. Grammar and spelling are corrected by our own twin model, trained on my texts. Sometimes a stumble stays in. That is mine too.

Read more All articles

Honest thinking.
Straight to your inbox.

One or two emails a month. No gloss, no spam.