Senior Dev: This "Grill Me" Prompt Is Going Viral Among Top Engineers

Okay, we're here in the AI Engineer Europe, uh, Podcast booth with Matt Pocock of AI Hero.

Hey, man. How you doing?

Good, man. Uh, it's nice to be in your neck of the woods for once.

Yeah. You do a lot of SF stuff, but-

A lot of SF stuff

... nice, so nice to have you come to London.

Yeah.

Nice to meet you, my man.

Um, how do we stack up v- versus your other London, uh, event experiences?

Yeah, it's the worst I've ever been to. Absolutely, by far the worst. Yeah, it's, uh, absolutely astonishing. Astonishing location, great stuff.

You and I first connected when you were about to start your sort of creator journey and-

Yeah

... your DevRel journey. Uh, you started Total TypeScript, which did very well, and then now you're, uh, running AI Hero. You are just about to finish a Cloud Code course.

Yep.

Um-

I run courses in, uh, two-week cohorts these days because AI moves so quickly.

Yeah.

Plan to release a self-paced course and guarantee updates forever, and so yeah, uh, on the last couple days of a two-week course. It's wonderful.

Does that mean that the Cloud Code source leaked just as you started?

Yeah. Absolutely.

How was that? What, what, what did, w- what did you...

Well, I'm cursed, you see- ... because I did a course on AISTK.

Uh-huh.

And, uh, it was based on AISTK V4. Uh, day two of the course, they announced AISTK V5.

V5, yeah.

Right? A huge breaking change. Um, and this time, uh, I think day two of the course, they released the one million context window, or maybe just before the course, uh, started. And yeah, now the Cloud Code, uh, source has leaked.

Um, but fortunately, the course is not really about Cloud Code. I noticed that when I was building it. It's mostly about engineering-

Okay

... and software fundamentals, and that's, I mean, that's the talk I'm giving here, too. It's kind of what I'm focused on now is going back to these old texts, figuring out what's good, and translating that into the AI era.

Yeah. So people can see the talk. You did a workshop as well. You're one of the... I had three people do a double session where, you know, because you're a big London figure-

Mm

... uh, I want to make sure that we have enough exposure to, to, to things. Um, and I think o- oftentimes conferences are afraid to double down on speakers, but I think, um, I have a particular thesis that YouTubers make particularly good conference speakers-

Yeah

... because you speak all the time for a living.

Yeah.

And, um, I don't know. Do, do, do you cringe when I call you a YouTuber? Is that a identity?

I'm used to it, right? I mean, like-

Yeah

... I mean, my identity is I'm a teacher, basically.

Yeah.

I really try... There's a whole category of YouTubers that I don't really, uh, try to interact with or, uh, try to be, which is the pundit, right? I'm not a pundit.

Ah.

I'm not a, uh, guy who's trying to predict the future. I'm just trying to teach how to do-

You don't, you don't change your favorite idea every two w- two weeks-

Exactly

... like some-

I'm-

... some other YouTubers.

Exactly. That stuff does do really well, right? And it's very... You know, it's hard not to fall into those traps because those numbers are really good. Uh, but yeah. I'm trying to teach stuff that's durable, that, uh, makes sense, that I feel good about.

Yeah, that kind of thing.

Yeah. Fundamentals and, of engineering and all that, what are the fundamentals that you hold dearly to?

I mean, I'll kinda rerun my talk a little bit now, which is like-

Yeah, a little preview.

Yeah, a little preview.

People can find it on-

Exactly

... on, online.

Yeah, yeah. The theory is that there's this idea that code doesn't matter anymore, right? We're in an era where you don't really need to think about the code or hold the code in mind when you're building applications because you can just use a description in English and use the AI as a kinda compiler to turn that into code.

You don't need to care about the code, it's just a compile target. And I think that every time I tried that, every time I tried to ignore the code, I would just end up with a, this terrible mess.

And so in kind of trying to better the compiler and improve the compiler, I went back to the old classic texts, you know, like Extreme Programming and Pragmatic Programmer, Philosophy of Software Design, and DDD, all that stuff. And what I found there was just, oh, I can just take these and just put them into my prompts, you know?

And I started going deeper and going deeper and going deeper and realizing, oh, okay, there's actually, there's so much more here than I realized, and that everything that we were doing for humans really works with AI, too. And actually keeping the code in mind and being really intentional about what you're building, not necessarily reviewing everything, but still keeping a sort of understanding of the architecture and the modules especially, it really yields dividends because if you have a code base that's easy to change for humans, it's gonna be easy for AI to change, too, and it's gonna mean that you can work a lot faster because a code base that's difficult to change for humans is, is gonna be even worse for AI.

Yeah. I've been thinking a lot about the concept of a narrow waist. I don't know if you've heard about this, uh-

No

... thing. It's, it, it, it comes from old internet architecture philosophy, but you've heard about maybe layer one, two, three, four, five, six, seven. Uh, the narrow waist is l- layers three and four, where basically TCP, IP, and HTTP have mostly dominated, and there are no other alternatives effectively apart from WebSockets.

Mm-hmm.

And I think introducing narrow waists in a code base allows you to contain slop, right?

Yeah.

You can say, "I don't care what goes on in here, but what goes in is defined by me, what goes out is defined by me. Do whatever you want inside."

Yes.

And so, um, some, there's some equivalence of this in my mind, of this and, uh, the former decompositions of MVC. Um, people call it model view controller. Today I call it model view claw. I view this a lot in, in terms of how I run AIE as well.

AIE is now a nine-person business. Everyone has different parts of the elephant, and I have to make sure that everyone sort of, um, uh, comes together and, like, produces a, a, a conference where nobody notices the, the, the, the rough edges down under, under the hood.

And so part of it is, uh, yes, I, I run the website. I run the speakers. I run the content, but everything has to flow from me down to, okay, well, what AV do you need? What hotels do you need?

What, uh, you know-

Mm

... uh, other, other sort of, um, accommodations, uh, do you need? And there's a huge coordination issue. Uh, it's just, it's just everywhere, and it, I don't even feel like it's really programming anymore. It's just systems.

Yeah.

Uh, y- in how you set up systems.

Yeah. This, I mean, what you said there about the idea of having self-contained boxes within, within any system, actually, but let's use a code base as an example.

Yeah.

It's a fascinating idea, and- John Asterhow in his book Philosophy of Software Design course talks about this, is this idea of deep modules, basically. A large amount of functionality with a simple interface.

Yeah.

Kinda ports and adapters, right? Like, you have a port that talks to other ports, et cetera. And you're totally right in that when you have a code base or a system designed like this, you can design the interface, be very intentional about that, and be a human sitting there and understanding it, but then delegate the implementation.

Right.

Delegate the stuff that's going on inside, and that's how I've used AI best for coding. I imagine for what you're doing, putting it into an org, that would work great, too.

Yeah. I see it at both ends, right? I, I, I also, uh, you know, consult for commission, and Devin at, uh, Goldman Sachs is, uh, very, very different when you have tens of thousands of engineers.

Yeah.

Um, a- a- and, uh, it's a, it's a beautiful thing to, to see. You mentioned domain-driven design, DDD. I am one of those people who I guess are old enough, uh, to, to have had that. We have actually a DDD talk, uh, yesterday at one of the workshops, uh, from Ikea.

Do you think it's catching on? Do you think it's important enough that I should be featuring it more? I think it's a very appealing concept for architects and people who work with large systems. Most people will never have encountered it in their software journey if they started more recently.

I don't know.

Yeah. It's a fascinating idea because what DDD is is a set of bu- Like, I'm not, like, a DDD expert. This is something I'm encountering relatively new.

But, but yeah, yeah.

But what you have with DDD is a set of building blocks and a set of practices that all fit together. And so you, what you have there, because it's been going around so long, it's kind of in the latent space of these, um, uh, these models already, is if you say you don't need to invent this whole new set of terms, you can just bolt on this framework, this system that's very, very flexible, very composable, and allows you to build systems, uh, in this way.

And because of that, it's very attractive. And so using a pre-built kind of framework for talking to the model and having the model understand exactly what you're saying because it understands, that just feels right. And because it's so focused on, um, outputs already and aligning the code with language, that's what DDD is, essentially.

You're trying to kinda create a unified domain model so that, the way I think about it now, is that so that the AI and you are speaking the same language.

That's so true.

That is so, so useful. So to make that concrete, I have a skill, um, I have a set of skills in mattpocock/skills, which got insanely popular, has, like, 13,000 stars or something. And one of those skills is ubiquitous language skill.

Ubiquitous language is essentially a concept where you create a document which essentially defines the terms of what you're talking about. And when you, uh... Let's say you're a client, and I'm a guy building a system for you. When I say, um, a mole, for instance, we might be talking about a health application where we're investigating moles, or we might be talking about spies or something, where we need to understand what a mole is in that context, you know?

Or the animal.

Or the animal, right? We might be building a zoo app or something, right? So you need all of these terminology, and anyone who's worked on any mature project knows that you have all of this arcane jargon that goes along with it.

So I have a skill that essentially scans your code base, looks for all of these terms, and refines them with you, and I notice that I keep that document open whenever I'm prompting because it means that I can-

Ah

... prompt so much more, with so much more clarity, and have so much more better conversations with the AI because it's using my domain model, using my language.

You keep the document, you keep the skill open?

I keep... I mean, it generates a markdown file-

I see

... that sits in my repo, and I just keep that open, look at it all the time.

Yes, yes, yes, yes.

So we're thinking along the same lines.

And does it, does that go in your agents.md, cloud.md, or is that literally just for your human reference?

I keep a reference to it in agents.md.

Yeah.

Um, I'm very nervous about putting too much in agent-

Too much, yeah

... yeah, but it knows it's there if it needs to see anything. But mostly when it's doing searches within the repo, when it's exploring, it'll use those terms and search for those terms, and the, the document comes up.

Yeah. Just, just more broadly on AI, uh, trends, I, I think you have, uh, been really good at explaining and teaching, uh, things. How do you do what you do? Your, your videos, uh, do so well.

Yeah.

Uh, I think the British accent doesn't hurt.

Yeah. Yeah.

But, uh, I think you're also, like, enormously amazing on Twitter, and, like, uh, you know, as much as people say I'm good at it, I'm not compared to what you do.

Well, I mean, I don't really understand it-

Yeah

... first of all. What I do know is that I spent six years as a voice coach before I-

Yes

... um, became a developer, and what I noticed when I became a developer is I would... I, you know, I started from the bottom, really. I was working as a junior, going up to mid, senior, blah, and the ability to communicate always just felt like a ridiculous overpowered skill that I had in my locker that no one else had.

And I think what I've chosen to do over the last couple of years is really narrow my focus. Again, I'm not trying to be a pundit, right? I'm just trying to teach stuff, and so that means I need to divide my time between learning the stuff and figuring out the right phrases to use about it.

This is why, again, going back to these texts are so useful 'cause I have all of these models now that I can use to explain things to people. Yeah, I don't know. I don't know why I'm so good at it.

Mostly, I don't feel good at it. Mostly, I feel like I just see the things I need to improve.

No, okay. May- maybe, maybe you can walk us through, maybe the CloudCore workshop or something else that you did. Uh, what goes into preparing something like that, right? Like, what's, what is the research process, and what do you throw out?

Yeah. Let's say we have an explore and exploit phase, right?

Sure.

Where, where I'm figuring out what this workshop might look like. My re- most recent course took about two months to put together. It's about, uh, I don't know, four and a half hours of very tightly edited video, but lots and lots of exercises, lots and lots of stuff in there, about 100 different units, let's say.

So let's say you, I figure out exact... Like, I start from scratch. What would be interesting to show? What's interesting in this environment? What's, what's my angle here? And I just make a ton of stuff inside an Obsidian vault, so tons of ideas, essentially a zettelkasten, if you've ever heard of that approach.

And these n- loose notes then kind of sort of start coagulating into a plan. And I have actually a custom application that I've built for planning this stuff too. So this means that I can then start seeing sections, start seeing groupings of information that I wanna see.

From there, I prioritize each individual lesson and work out how important it is to the running of the course, P1, P2, P3. I then record all the P1s mostly, and then that's the course essentially, 'cause most loads of stuff gets left on the cutting room floor 'cause we just can't fit it in, or it has various concerns and things.

So that approach in developing this idea that each lesson should only teach one thing, each lesson should depend, and the dependencies between all the knowledge should be super clear, and you should always challenge people without overwhelming them. And this is, I suppose, just instincts I've built up over a long time of doing this.

But if you keep that idea in your head, then it's hard to go wrong.

Yeah. Do you feel any changes in how people learn?

Yeah.

Because there's a lot of... You can just ask Grok, you can ask ChatGPT. I don't know. Is- this is a- a- I guess, a more broader education concept, right, of, um, w- like, do you g- give people more exercises?

You know, is, is it more hands-on?

So there are three things that you can teach people. You can teach them knowledge, and you do that through lectures, right? This is how things work. These are the terms. You can teach them skills. Y- this is how to do it.

Try doing it with an interactive exercise. Or you can teach them wisdom, and wisdom is the hardest. How do you teach wisdom? For me, that's been, um, doing discussions. Small group discussions and workshops is kind of the recommended approach.

But wisdom is something that is really hard to get. And actually, working with AI is something you can, can sort of get to wisdom a little bit by talking to AI. You've never really be able to, able to have that with any kind of, uh, learning mechanism before.

You know, books are mostly knowledge-based, you know. There's a little bit of wisdom you can glean from them, but you've gotta do a bit of work on your own to get there. But what I've noticed, the more I lean into the kind of AI experimental stuff, the more it actually turns people off my materials.

I'm finding that actually a traditional approach is w- w- what w- what still-

Give me the lectures. Yes

... yeah, what still works for most people, to be honest.

Yeah. Yeah, yeah.

Most people just still wanna learn a traditional way, and I'm totally fine with facilitating.

The, uh, you know, back when there was the creator economy wave, and Maven was all about the interactive discussions, and it turns out people just want lectures. Um, and-

That's right

... they, they... Maybe a bit of homework, but a lot of them don't even do the homework.

That's right. Yeah. And I-

So, so what you're saying is that I've arrived at the optimal form factor, which is just, just lectures.

I mean, I- I'm very careful with my stuff to, um, really force people to do the work essentially.

I see.

E- even in my TypeScript stuff, I would, uh, structure it so that people would have almost... Uh, you would essentially just be thrust into a problem and have to solve it on your own, and I would give you the knowledge afterwards.

And that approach has felt pretty good really.

Yeah.

So, yeah.

Something I've noticed, when I started AI Engineer, I was pretty conscious about it being 50% Python, 50% TypeScript. I would say that's changed. I think this year is the year that TypeScript overtook Python in the GitHub survey, and, um, I didn't really foresee that coming.

I think that Python is somewhat more expressive, especially, uh, useful in the back end. And TypeScript has the TypeScript dependency issues that we've always had in TypeScript. I don't know. I, I think, like, uh... Do, do you think TypeScript's winning?

Yeah.

I mean, you, you would be biased towards saying yes.

Of course.

Yeah.

And, um, my echo chamber is 100% TypeScript, and always has been. And again, no punditry. I, like... I'm not a pundit. I mean, what you have with TypeScript is this incredibly rich ecosystem of frameworks and tools and, uh, extraordinary amount of-

You, you have the cl- Vercel Next.js and the Cloudflare Next.js. Uh.

Exactly. Right.

I'm not... I'm, I'm, I'm taking the piss. Uh.

But, you know, y- y- so you can't overlook TypeScript in any conversation, you know, uh, when it comes to, uh, building applications. And that just seems like everyone who's building this stuff, everyone who's building chat applications-

Yeah

... is mostly doing it. If you're concerned about UX, you're concerned about shipping great stuff, you're mostly doing it in TypeScript. At least, that's what I see.

Yeah, yeah. Um, I, I, I think it, it could be that TypeScript ha- is gonna win AI engineering, and that's something I haven't anticipated or seen, or I don't even know how to sort of position around this. Because it- I think it really does mean a lot of different things for, um, what frameworks I choose to, to promote or, uh, bet on, uh, in, in, in my line of work.

Uh, yeah, it is interesting. Any other, um, topics in AI that you're particularly captivated by? You know, you're, you're finishing up your cloud code exploration. What's next?

Well, I think this has been the most successful course I've ever done. So, like, for the last 18 months, I've been trying to figure out my positioning, trying to work out where I fit. Um, I even sort of...

That idea of sort of TypeScript coming and eating Python is something that I was betting on for a little bit. I even made a, a TypeScript evals framework called, uh, Evalite, uh, which I've been trying to-

Oh, yeah. I didn't see that. Yeah

... yeah, which has sort of stalled in its development and stuff. Um, but, you know, people are not really interested in evals. You know, you... Like, evals are not sexy. Like, no one's excited to do evals these days, right?

We have many, many people who would disagree here.

I mean, from everyone I've talked to and the content I produce on it, it do- doesn't go anywhere, or at least people-

Yeah

... aren't kind of into it yet, I think. Um, in terms of where I'm going next, uh, I think that, I think that the coding agents kind of debate is sort of ending really. Like, they're mostly coagulating towards a similar approach, which is, you know, file-based systems, a simple set of tools.

And I think I'm interested in expanding into that kind of area and looking at, uh, how to make a course that potentially, like, does all the agents at once possibly and sort of trying to build frameworks and build methods of, uh, you know, software factories that are kind of independent of the agent that you're using.

Yeah.

So a kind of meta level on top of that.

Yeah.

That's kind of where I'm at. And I think that coming up for the next couple of, maybe the next few months, let's say, I'm not a, not a, a clairvoyant- That the inversion of control is gonna be really important, where you put more control in the hands of the developer and less in the harness.

Yeah. Um, oh, I thought the developer wants less control, not more. Is that not true? I think there's a trade-off, right? Because if you're using Claw Code, let's say, and, uh, until recently you couldn't see Claw Code's internals, right?

Mm. You sort of had a, um... There was a sense of a box around Claw Code that you didn't need to look inside, and that's comforting, right? Because you just sort of go, "Okay, what, what does the harness want me to do?

Let's lean in its direction. Let's not worry too much about what's inside." And what you get is you get ease of use, but you lose observability- Mm ... and you lose the ability to kinda control what's going on and tweak what's going on.

And there's... That trade-off is really interesting to me because let's say we look at Pie instead, which is composed out of tiny little primitives that you totally build up from the ground. Now, what that means if you're able to use Pie really well is you have total control over the system, and you can observe all the parts of the system.

But then you have more stuff to maintain, stuff to care about, stuff that can go wrong, and it's all up to you. So that trade-off is something I'm interested in exploring next, for sure. Yeah, I think it's very interesting, all the, these, these developments.

I think the world needs more, like, teachers like yourself. Um, I have been also asked to do more stuff like that on, on my podcast, and a lot of people are trying to also just, like, get out... get more productive with AI, and I wonder how to do that well.

Um, I think you're obviously more focused on engineering and programming and software engineering, uh, but there's a, there's equivalents of, like, more generalists that want to know, "Okay, well, how do I live my life more efficiently, just in general?

How do I run my team and my company more efficiently?" Totally. It seems like Obsidian is, like, kind of winning there, weirdly. Mm-hmm. They're four people. Very crazy. Really? They're, really, that small? Four people. Yes. Yeah. Wow, I didn't know that.

They, they, they recently posted a job for a fifth person- ... and that thing got nine million views- Wow ... because everyone was like, "What do you mean you, you- That's amazing ... you've done this with four people?"

That's crazy. I never knew. Wow. Yeah. I mean, that... The stuff that we're doing as software engineers now is like we seem to be a step ahead of everyone else because we're able to work in a domain where AI is actually producing quality stuff.

And so what I'm doing is as I'm working with my, with my set of skills that I'm kind of authoring, I'm thinking, "How can I pull these out and use these in my everyday life?" For inst- what do I take out of Claw Code and put- I'm gonna look up your skills right now, actually ...

in claw.ai. Yeah. Yeah. Matt Pocock forward slash skills. Mm. And there's one, one in particular called Grill Me- Okay ... which is like three, uh, sentences long- Okay ... and which basically gets the AI to interview you relentlessly, right?

Until you reach a shared idea. I'm sure everyone sort of had that technique or used that technique. And I've realized I need that skill in all sorts of domains. Any kind of domain where I want the AI to be aligned with me, I want to reach a shared understanding with the AI.

Yeah, yeah, yeah. And I want it to sort of artificially maybe even sort of just question its own assumptions, question my assumptions until we reach a shared design concept, and I've used that for all sorts of stuff, for generating documents, for doing tons and tons of things like this.

And so I think the more I break these down into individual composable parts, the more useful they are outside of software engineering, too. Yeah. Just like the 20 questions and, and all that. Yeah. Um, okay. This has been beautiful to catching up.

Uh, I w- also want to intro- introduce the Lead Space audience to your work. Um, I think the work of explaining AI, keeping up to date is never done, so, uh, just glad to have friends and, uh, who've also c- made the crossover from JavaScript to AI.

Absolutely. Yeah. It's, uh, been a fun journey. Yeah. Continues to be. All right. Thank you. Thanks.