Resources

Article

Why founders need to get their hands dirty with AI

From building personal bots that monitor competitors to using deep research tools for board presentations – Michiel Chevalier, Founder and MD of Nmbrs, proves that the only way to truly understand the ROI of AI is to put in the hours yourself. We sit down with Michiel to discuss his approach to AI, why renting an expert is a mistake, and the only human skill that matters in an automated world.

Article

Why founders need to get their hands dirty with AI

From building personal bots that monitor competitors to using deep research tools for board presentations – Michiel Chevalier, Founder and MD of Nmbrs, proves that the only way to truly understand the ROI of AI is to put in the hours yourself. We sit down with Michiel to discuss his approach to AI, why renting an expert is a mistake, and the only human skill that matters in an automated world.

Article

Why founders need to get their hands dirty with AI

From building personal bots that monitor competitors to using deep research tools for board presentations – Michiel Chevalier, Founder and MD of Nmbrs, proves that the only way to truly understand the ROI of AI is to put in the hours yourself. We sit down with Michiel to discuss his approach to AI, why renting an expert is a mistake, and the only human skill that matters in an automated world.

AI, Business insights

Article

Why founders need to get their hands dirty with AI

From building personal bots that monitor competitors to using deep research tools for board presentations – Michiel Chevalier, Founder and MD of Nmbrs, proves that the only way to truly understand the ROI of AI is to put in the hours yourself. We sit down with Michiel to discuss his approach to AI, why renting an expert is a mistake, and the only human skill that matters in an automated world.

AI, Business insights

First, tell us a little bit about Nmbrs.
Nmbrs is an HR and payroll platform suitable for SMBs. We are mainly in the Dutch market, where we have around 40% market share in SMB companies.

We built this ground up to be a cloud-first HR and payroll platform. The old-school payroll platforms were really dedicated to the payroll professional, but we took into account that it had to be a collaboration tool between the payroll professional, the HR manager, and the employee. Our story was always: if the HR process is automated, payroll does not have to work to get all the information they need.


What is the one specific task, business or personal, where AI saves you the most time right now?
For me personally, I use it for all my communication purposes. Both on social platforms like LinkedIn, but also internally for sending emails and recaps.

Next to that, I also have a few agents I built myself to monitor competitors, get signals, and generate monthly reports. I also use it for market analysis. When we want to build a new plan or strategy, we use deep research tools to bring business cases to our board. I do all of it myself, because I’m trying to make my life as easy as possible.


Can you share one concrete example where your company used AI to add real value for the customer?
We have two streams of AI within our company. One is to facilitate customer flows, like support and sales, to help customers get real-time answers. For example, we launched knowledge assistance about a year and a half ago.

The second is in-product algorithms, where we use the technology to help them do their tasks more easily. We launched a salary assistant so that they can help fill out new mutation forms or get reports.


How do you measure if an AI tool is actually providing ROI?
The most important measurement at this moment is the usage of the AI tooling we're offering to our customers. How much is this AI agent being used, by what type of users, and what are they using it for?

What we see is that the return on investment is not just that they can do something faster on our platform. The biggest ROI is if we can really help them do a job that they were currently not able to do. Currently, the focus of our agents is actually helping customers with things they are doing outside of our platform.


How do you distinguish between AI hype and tools that bring value to your company?
We do AI experiments. We’ve always been experimenting, ever since 2016, building payroll predictors and such.

Only recently, when we see something really big that is picked up by our customers or has big value, do we do it on a bigger scale. For instance, the MCP (Model Context Protocol) framework launched by Claude – it became clear to us that this was something we as a software provider had to embrace. This was one of those moments we invested heavily in.


As technology gets smarter, what human skill matters most?
Genuine curiosity.


What is the biggest mistake you see other companies making when rushing into AI?
Overpromising and underdelivering. The pressure to ride the hype is big, but you have to bring your customers along on the journey. You have to know not just what customers are doing on your platform, but also outside of it. In what ecosystem? It’s easy to over-invest without really knowing where the value is for your customer.


If a founder asked you where to start with AI tomorrow, what is the very first step you would tell them to take?
Do it yourself. Just go all in.

There is this rule, the 10,000-hour rule, if you put that effort in, you become a specialist. I think every founder should put this effort into really seeing business value and where to invest. Make sure you become the expert instead of renting an expert. As a founder, you should understand it.

Related content