Built with you, run by you
Your know-how,
running as a system.
We turn work that lives in people’s heads into AI systems, and teach your team to run them.
How we work
From know-how to a running system.
Every engagement runs the same three steps. The same approach built the Nitai Suite.
Understand the work
We sit with the people who do the job and map how the work really flows: the decisions, the exceptions, the tools.
Build the system
Specialist agents on your own data, sequenced by one orchestrator, with a person reviewing the work at every step.
Hand it over
We teach your team to run it. You own your deployed instance, and every correction your team makes improves it.
Where to start
Two ways to start.
Whether you have a workflow that should run itself or a need one of our products already covers, the first conversation is about how your work happens today.
01
Build a system around your work
For work that lives in people’s heads and scattered tools.
Multi-agent AI designed from your team’s workflow and data, with human review built in, handed over for your team to run.
- Discovery
- The system
- Your team, trained
02
Use one of our products
For when one of our products already covers the need.
Products for content in every language you sell in. Use one, or several together, each as your own white-label instance.
- Multilingual content engine
- Editorial operating system
- Video dubbing
- Audio Studio
- Company Brain
In production today
The Nitai Suite
Products that each do one job. Use one, or several together. A person checks every result before it goes out, and the products run inside WordPress or other platforms, as a standalone app, or through integrations with the tools you already use.
Multilingual content engine
What you have published, translated, and turned into posts for every channel and language.
Editorial operating system
Seven agents take a piece from the first market signal to your site, email, YouTube and social.
Company Brain
Everything you have published and every correction your team makes, for the other products to draw on.
Video dubbing
Your videos in another language, with a real voice for each speaker.
Audio Studio
Your articles as narration, with a listen button beside the written piece.
800+
articles and dubbed videos
from Gear Tips’ existing Portuguese material, translated and dubbed, every one with its full revision history
Any
language
your readers need. Gear Tips publishes in English, Portuguese and Spanish at the same time
Gear Tips uses it every day, and has asked for the next version.
Why it gets better
Every review improves the next result.
The system’s first version, your team’s correction and the approved result are all kept. Your team’s own corrections become the data that makes the system sharper.
01
Draft
The system makes a first version from your own material.
02
Review
Someone on your team corrects it, and says why.
03
Record
The correction is kept, with the reason attached.
04
Improve
The next result starts from the corrected version.
Every turn runs the same four steps, and every turn needs less correcting.
A correction your team made, kept with its reason
Your standard: what your team would approve without a change
An illustration of how it learns, not data.
The team
Three people, no overlap.
Nitai Labs is an AI consulting practice based in Barcelona, run by two founders and a founding architect. We work in English, Spanish, Portuguese and Dutch, with clients in Europe, Brazil and the USA.
Founder · CRO
Paula Coelho
25+ years in enterprise sales and growth. Former Microsoft developer and co-founder of Vesta Technologies, a B2B gateway company that reached a $100M valuation. Owns clients, revenue and positioning.
Co-founder · creative director
Iris Heddes
Trained in architecture, with a postgraduate in multimedia. Runs product design, operations and workflow orchestration, and quality in Dutch, English, Portuguese and Spanish.
CTO
Johan Kruger
Principal systems and agentic architect. Designed and built the multi-agent orchestrator behind the Nitai Suite and most of its products, and owns the technical roadmap.
In development
What we’re building next
Our own models for the Suite, trained on your team’s corrections, and Virtual Stylist, a platform for stylists and their clients.
Questions
What people ask before the first call.
If yours is not here, ask it.
What happens in the first conversation with Nitai Labs?
We ask how the work happens today: who does what, in which tool, and where it stalls. It takes thirty minutes and costs nothing. If it helps, we walk you through our deck and show you work we have done for clients in a similar situation. If there is a system worth building, the next step is a discovery of up to four weeks, and you know its price before it starts.
What does working with Nitai Labs cost?
The first conversation is free. Discovery is a fixed price, set before it starts, and half of it is credited against the build if you go ahead within 30 days. The build is priced from what discovery finds, so you see the full figure before you commit to it. The products are priced by the modules you run and the volume.
Do we have to change our tools to work with Nitai Labs?
No. The system is platform agnostic: it runs inside WordPress or other platforms, as a standalone app, or through integrations, so we build around the tools you already use. If something does need replacing, we explain why, what it costs and what keeping it would mean. That decision stays with you.
Where does our data go when we use a Nitai system?
Your data stays in your own instance, with your team’s corrections. The system runs on your own accounts with the model and voice providers. Only the text or sound being worked on is sent to them.
Who owns the AI system Nitai Labs builds for us?
You own your deployed instance: the system as it runs for you, with your data and every correction your team has made. The Nitai Suite underneath it stays with us, which is why each client starts on a core that is already in production rather than a new build.
What happens when the AI gets something wrong?
Someone on your team corrects it before it goes out, because every stage returns to a person before the next one runs. The correction is kept with the reason for it, so the next draft starts from the corrected version. The review step is what makes the output publishable, and what teaches the system.
Does Nitai Labs work in languages other than English?
Yes. We work in English, Spanish, Portuguese and Dutch. The multilingual content engine translates published material, narrates it and dubs video line by line, with a voice for each speaker, so one source reaches readers in any language you need.
Start with how your
work happens today.
A free thirty-minute call. We ask who does what, in which tool, and where it stalls.