We sat down with our co-founder, Dan, to find out about his background, experiences, and what he’s learned in 20 years of machine vision.
Behind every Cortha project is a team of experienced engineers, project managers, and machine vision specialists. In this series, we’re getting to know the people behind the work, from our co-founders to our latest recruits. We’re starting with our co-founder, Dan Maynard.
Tell us a little bit about your background.
“I did a degree which was mostly mechanical. Then I did a postgraduate course where I visited lots of companies and did lots of projects up and down the country. And then I worked for an engineering consultancy in Cambridge, in general development engineering in the field of medical devices. I’ve always liked a bit of building things… where you turn something on and then the machine does something that makes something else happen.
Then I went and worked in an automation company which builds special-purpose automation. It was a small company, so the job involved doing mechanical design, bits of assembly, bits of wiring, the software, and supporting things in the factory. So it was a good learning ground and I was there nearly 10 years. Then I decided it was time to move on, and I set up my own business doing robotics and motion control and vision, but I became so busy doing machine vision that I’ve done some robotics but very little motion control. I’ve basically been doing machine vision for years and years.”
Was that down to higher demand?
“You get tipping points for technologies. There are a number of things which are good enablers; digital camera technology, PCs being stable enough to run in factories, these sorts of things made it possible. But also, the general trend has been that the cost of technology has gone down and the demand for quality has gone up, so it’s an expanding market.”
What lessons would you say you’ve learnt over the last 20 years?
“20 years is a long time and a lot of lessons! I’ve learned all kinds of stuff about machine vision algorithms, optimisation, how to programme things. But there were some things said to me early which I still think are true. A lecturer once said, ‘Before you begin any project, you have to know what you’re going to do, how you’re going to do it, why you’re doing it, and how you’ll know when you’ve finished.’ Then he gave us a list of examples of projects where people have not known one of those things, and it all went completely wrong.
If you don’t know those four things, then you’ll either come a cropper at the beginning, in the middle, or at the end of the project. That was definitely one of the most valuable things that came out of the course.
Another thing I learnt is people do business with people – and that’s still true. Years later, even internet-based, it’s still true. And I’ve always had a strong ethic of trying to do things properly before we ship them, otherwise it causes more trouble down the line. Also, you should do things you’re good at and find other people to do the things you’re less good at!”
I think every business owner has learnt that at some point! When did that appear true to you?
“I was once on a training course with a guy who said to me, ‘You’re going to get people to do things, but you have to delegate and accept that they’re only going to do things 80% as well as you would have done it if you’d done it yourself.’
I think that’s sort of true, but what you actually have to do is accept that some of the things might be done 80% of the way, but look out for things where they’ve done it better than you would have done – or where they’ve had an idea that you wouldn’t have had. Because those are the gems. Those are the real things you need to celebrate and encourage, because that’s how you become stronger team, isn’t it? You have to accept that you don’t have all the answers.”
What are some common mistakes you see other companies or engineers making?
“I have seen systems where they haven’t thought enough about, say, lighting. And in a machine vision world, or what used to be called image processing, there are lots of tasks like facial recognition or counting number plates on cars, for example, that become optimised for the industry using it. In machine vision, we’re inspecting things in factories or controlled environments, doing the same task over and over. If you’re recognising people walking into a casino, you just have to deal with whatever the lighting is like and wherever the people are. But in a factory, we’re much more able to constrain the situation. In the casino, if you recognise 90% of the people, you’re doing pretty well. But in a factory, that’s not good enough.”
Is there anything you personally focus on to make sure that Cortha’s systems are reliable in the real world and not just in theory?
“I’m a big believer in understanding how things work properly. That’s how you see what’s affecting it and then when it stops working, you know something changed and you can work that out.
Also, you shouldn’t try and reinvent the wheel. Where people have already built good quality components to do specific jobs, you should buy them. Maybe the purchase price is a little bit higher, but they’re more reliable, so you should always buy reliable components.”
What would you say makes your partnership with Paul so strong?
“Paul and I go back a long way. On a personal level, we get on. And on a professional level, we complement each other. We come up with totally different ideas and it’s great. For example, would I have thought of doing this interview? No – yet here we are!”
What excites you most about the future of machine vision and where Cortha is heading?
“I think the future of machine vision is interesting. People invent things and try and sell them. So I look at all these things that come along, and I ask, do we need it? Or, is it better? And the truth is, sometimes it’s really useful and you do need it, but most of the time, you don’t.
But I think there is a tipping point with neural networks, and it’s because of increasing processing power and involvement of some of the big players like Facebook, Meta and Google getting behind it. So there’s a place to implement that technology in machine vision. The invention of deep learning has got legs and there will be brilliant applications from it.”
Dan brings not just decades of technical expertise to Cortha, but a clear philosophy: build machines that work, build trust with people, and never stop learning. It’s that combination of curiosity, rigour, and collaboration that shapes how we approach every project.
Interested in what Cortha could do for your business? Contact us to start the conversation.