Anyone who has built a working prototype with ChatGPT, Claude Code or Codex has already proven the hard part: that the idea solves a real problem. We take that app to production with authentication, real data and integration, and the code stays in your company's hands.
More and more teams are building their own apps with ChatGPT, Claude Code, Codex or Gemini, and that is good news. Instead of taking a decision maker a document proposing an improvement, they take the process working, with data in front of them and results in plain sight. It is the fastest way there is today to validate a business idea, and anyone doing it is doing the right work.
What comes next is not rewriting what already works. It is the layer no prototype has, because that is not what a prototype is for: authentication and permissions by profile, real data instead of test data, a link to the ERP or the CRM, backups, tests, and a record of what happened when someone reports an error. We take what already exists and add that layer, with the code, the database and the infrastructure ending up in your company's hands.
The presentation worked, management gave the green light, and from there the questions changed in nature. Who is allowed in? Where does the data sit? What happens if this fails on a Tuesday morning? Who answers if one client shows up on another client's screen? These are fair questions, and none of them is about the idea, which is already proven. They are about the foundation the app needs in order to be used across the company, and that foundation does not get built in the same session as the prototype.
Nothing starts over. What already works gets the pieces it is missing so that ten, fifty or two hundred people can use the app every day without surprises.
Authentication tied to the Microsoft 365 or Google accounts the company already uses, with permissions by profile and by area. Each person sees what concerns them, and there is a record of who viewed or changed what.
A database of your own, with versioned migrations so the model can change without breaking what is already there. Whatever exists in the prototype is migrated, and backups are tested with an actual restore rather than just configured.
The app stops working off manual exports and starts reading from and writing to the systems where the data lives, with error handling for when the system on the other side does not respond.
The keys that ended up inside the code during the prototype move into a secrets manager, and any that have been passed around get replaced. We add monitoring, usage logging, automated tests on the flows that cannot fail, and a test environment separate from the live one.
An app in production is only worth as much as its connections. We connect to what the company already has, including whatever runs on your own servers with no modern API, and we keep the tools the prototype already uses whenever that makes sense.
The app was born inside your company and that is where it goes back. It does not end up dependent on us, on accounts of ours, or on one person being the only one who knows how it works underneath.
It starts with what already exists. The first thing we do is watch the app working as it is, instead of designing another one on paper.
We watch the app working, see who uses it today and what has already been validated. You leave knowing what is missing before it can go into production, and in what order. No commitment to proceed.
We go through the code, the data model, the integrations and the credentials in place. At the end, scope, timeline and investment are fixed, with what gets reused and what gets added identified point by point.
The app starts running on your company's infrastructure, with authentication, permissions, backups, usage logging and one real integration. The people already using it keep working through the switch.
More teams, more modules, links to the remaining systems, automated tests on the critical flows and whatever compliance requirements the company has. From there, the app evolves at the pace of the business.
Reference bands upfront. The exact figure is fixed in the proposal, after the technical review, and does not change midway. The diagnosis is free and comes with no commitment.
€7,500 to €15,000
The app you already have starts running on company infrastructure, with authentication, permissions by profile, backups, usage logging and one real integration. Includes reviewing the credentials that almost always ended up inside the code. 4 to 8 weeks.
€15,000 to €35,000
Data model stabilised with migrations, a link to the ERP, the CRM or the production database, migration of what is already in the prototype, automated tests on the flows that cannot fail, monitoring and a test environment separate from the live one. 8 to 14 weeks.
From €35,000
The first 3 months of support are included in the bands above. After that, extending to more teams and more modules, tougher compliance requirements, volume and availability.
Free diagnosis, no commitment. We look at your process and come back with a plan covering scope, timeline and investment. Reply within 48 business hours, NDA included by default.
Two minutes to tell us the essentials. We reply within 48 business hours with a first read on your case and, if it makes sense, a proposed session.
Every vertical has its own challenges. See how we approach other sectors where we have proven experience.