Welcome. So, this workshop is on Ralph loops. Uh, hands up here who knows what a Ralph loop is. That's almost everyone... What we're going to do today... we're actually going to build Ralph Loops together... Hopefully the idea is that you'll be able to use this on your real work when we get done.

Who is using Claude code or codeex specifically to write code? Hands up... Who is no longer writing any code? That's quite a lot of people. Look around for a minute. That that is a huge change... My name is Chris Parsons. Um, these days I spend most of my time trying to help, uh, teams... figure out what on earth to do with AI mostly. So, uh, I'm a CTO by background... I have about 30 years or so of building software professionally... and funnily enough all of those uh principles and practices that we taught for years [Agile]... are still very much applicable um to AI.

This is how I used to work with AI until quite recently... This is an n8n workflow uh that I used in order to create my weekly newsletter. It took me uh probably a week to write let alone actually test and debug... it was so brittle to use at this kind of level of complexity... And I thought that this was the future of automation, but it isn't really the future of automation. The future of automation is a lot more like something like this uh running in Claude code... I have now in Claude code a skill that writes newsletters for me and it has all of those instructions. In fact, I copied and pasted the n8n JSON code from there into Claude code and said, "Write a skill based on this flow, and it did a great job."

Ralph loops Fundamentally is running on a loop, isn't it? It just reads the skill, calls a tool, goes back to the beginning, reads the skill again... at some point it figures out that it's done and it stops... The first idea came from Jeffrey Huntley uh a little while ago... ancient times in AI which means probably about last June and he said basically what we should do is whenever we finish using an AI to do anything we should just try the thing again... give it exactly the same prompt and see what happens...

It's called a Ralph loop because of Ralph Wigum, which is a Simpsons character who basically says um the uh he just tries the same thing over and over and over again and eventually it works... All that a Ralph loop is is uh build this thing or do this thing inside a prompt. Then the AI goes away and does the thing... and then you give it another prompt again say go away and build the feature... oh yeah there was actually this tiny thing that I should have done I really am now finished and so on and so on... Where you just build the feature and just then ask it to build the feature and then you ask it to build the feature.

The really interesting thing is not how do I make sure Claude has finished this one thing. It's what happens if I point this kind of loop at a whole pile of things to do, right? What happens when we point at a whole list of things?... what I'd done effectively was recreate the waterfall processes... where the entire project was specified up front... No wonder it didn't work. If humans can't do that, how is AI supposed to do any better?

However, if instead of saying with all of your tickets... you can just run a loop where you say something like, "Hey, just pick the next most important ticket." It's as simple as that... the AI is quite capable of looking at all of them, figuring out the dependencies on the fly based on what's just been done... let's forget parallelism just for a minute and just start with a loop... If I go back to Claude... and say implement the next most important ticket using uh TDD principles from doc tickets. um commit when done.

I can just simply go into Claude and say implement the next most important ticket... now the interesting thing now is that once this is finishes, now it's using TDD, so it read the test first... You could have a loop that does something like this: loop every one hour check linear for new bug reports from test... I have a skill that I'm working on... which is called startup... the idea is that it should um basically guide a product through an entire startup framework... it is meant to be run as a loop.

The reason for showing that is more to kind of point out that that AI can do a heck of a lot... I have a morning loop where every morning at at 6:00 a.m. it comes up with a full briefing of my day... I have a worker loop... It's got a decision trail of things that it's done and why it's done them... My basic rule is: is this reversible without embarrassment to me? And um if the answer is no, don't do it.

There's a thing called the lethal trifecta... Simon Wilson coined it. um an idea that if you have untrusted tokens, uh internet access and access to secret important data you don't want to lose, you're going to lose that data... I use a combination of uh positioning the code physically away from the machine on a VPS... I use Claude permissions... Use sub agents here now for the validation step, it started finding things...

I'm always a bottleneck. I have 30 different things that I need to now review... The only reason that it's given it to me is because I can't trust the AI with it... Within any team in any system there is always a bottleneck... This is why some teams when using AI tools actually go slower... because they're not working on the constraint... Always fix the thing that is the biggest bottleneck first... If you want to transform your team, you are going to have to sponsor these kind of experiments and be okay with failure... give um your team space to try a whole bunch of different things. So give them air cover.