So when you've understood what the problem is and how do you then get the right resources in place? Do you have to have different team members with different skills? Do you have to have different agents? Do you have to have different technologies? yeah

so i think there are some patterns so that we we break this down into horizontal and vertical patterns in a business so there are some challenges that every business has like sales you know it's a really good one like back office invoice um query response you know there's there's horizontal ones so in our library our blueprint we call it for every industry we have horizontal I'm going to use the word agents very loosely because it's a word that people use.

It's not quite the right word.

Then there's vertical specific ones.

In construction, we've got a concrete, a crush workflow testing agent because that's a thing that happens only in construction.

And I think the reason we picked three verticals is we can be very specific about those blueprints.

So we are saying, here's 100.

They're

Why is that distinction so important? And what goes wrong when companies automate workflows without putting governance guardrails in place?

On a sort of basic level, if you're a prosumer or consumer user of, say, Claude or whatever, ChatGPT, you can afford for it to do certain things you don't expect.

You can ask it to create a document or an output or an agenda or something, and it can give a different answer every time.

And that's okay.

At consumer level, you can deal with that.

Enterprises, whether it's small or large companies, they need certainty of outcome.

I think that's number one.

But what does that inflection look like for mid-market companies specifically? Where do you see the practical AI transformation landscape in, let's say, three to five years? And what do you believe Neologic needs to get right today to be the platform that defines how that segment makes