Dylan Ratcliffe

I'd thought that if we could discover every dependency in real time with no input from the user, because of course, if you expect the user to tell you where to look, then you're not going to find the things that are going to cause the actual problems, then this would solve the problem.

And so we built a real-time discovery engine that could do exactly that, find every dependency within your cloud environment and show it to you.

It took about six or eight months.

For the backend, I wrote it all in Go, not because I knew how to write Go.

I didn't.

I had to learn how to write Go.

But because I knew that I would be able to hire infrastructure people who knew Go.

My name is Dylan Ratcliffe, and I'm the founder and CEO of Overmind.

As a company, our mission is to remove the dread that people feel when they have to press the button to deploy to production.

We do that by simulating your infrastructure changes against your real-time actual production infrastructure before you make those changes and therefore preventing any outages before they happen.

I started building it because I'd seen this dread before.

I'd experienced the dread before because very often when you're deploying, especially infrastructure changes, which is what my background is and what we focus on, things can go wrong for reasons that are entirely not your fault.

And being afraid to press that deploy button, especially if you work in a big company with big complex infrastructure, which is built by people who've since left, being afraid to press that button is like a good survival instinct.

the thing that actually got me started in quitting my job was a particular incident with a printer so i was working at a financial services company in london which will remain nameless and we were deploying puppet we had a week to do it we'd done everything we needed to do we wanted to get a big win in before the end of the week and so we had this change in the pipeline ready to go

which was instead of having one password for all of their systems which i realize is terrible and that's why we're getting rid of it people would instead log in with their username and password and so this is like a linux backend server fleet so you use triple sd to set all this up and so this change would move from there being one password which was like written down in sticky notes around the office to actually people's usernames and passwords so when they

quit the company we could delete them and all kinds of stuff and we had done all the testing we tested it in the dev environment we got all the approvals it was absolutely ready to go and we were ready at 3pm on Friday to press the button and just get a big win in before the end of the engagement the 3pm on Friday thing should be foreshadowing to anyone who's ever worked with infrastructure it's a bad idea to press that button at 3pm on a Friday especially when you're supposed to turn into a pumpkin at 5pm because

The engagement is over and they're no longer paying you.

But we did it anyway because we were legitimately confident.

And the phone rang.

Well, it didn't ring initially.