Latent Space: The AI Engineer Podcast
00:41:33 · S2, Mihail Eric, Shawn Wang
[00:41:33] S2: Seems legit. CLI is a fun one. Do you have a lot of students, or this is their first time in a terminal in a CLI? Or is it kind of the norm now? [00:41:41] S1: No. I don't think so. I think that given that most of them, again, last year, I would say that some of them had probably still remember the days of finger on keyboard typing. This is, Most, I think, students have interacted with Claude Code at this point. [00:41:57] S2: I'm curious the split of people that are now just using the apps, right? Claude Code also has a desktop app. So does Codex. Codex is primarily an app now. You're kind of not in the CLI, not in the terminal. It's a thing that puts off a lot of average people, but just random note. [00:42:17] S0: Yeah. I would say, you know, I think there's also another thing about skill. There are... That are bundles of skills where they they sort of cross reference each other. Yeah. [00:42:25] S1: Like the plugins. Right? [00:42:26] S0: Like the plugins. And then there's also, I guess, the art of writing a skill, which is a thing that I've also focused on very IE. Mhmm. Mhmm. I think a lot of these are focusing on skill consuming. Yep. And you have all these influencers like Matt Pocock and what have you. They're like, yeah. This skill will save you. It will solve all your problems. But, no, you should write your skills yourself. [00:42:46] S1: Right. [00:42:46] S0: And, actually, when you try to write them naively, they're usually bad. And, actually, it takes a little bit of skill to write a skill.
00:42:46 · Shawn Wang, Mihail Eric, S2
[00:42:46] S0: And, actually, when you try to write them naively, they're usually bad. And, actually, it takes a little bit of skill to write a skill. [00:42:55] S1: That's absolutely right. That's absolutely right. [00:42:56] S2: What are swyx's best tips on writing skills? [00:42:59] S0: I have a skill... I have a public skills repo. I went through definitely a phase of psychosis of, like, skill everything. [00:43:03] S2: No. No. Don't give us skills. How do you write good skills? [00:43:06] S0: Yeah. Yeah. Exactly. So I have a skill for cutting skills. So I I I... Like, there... So every, like, Claude Code and Codex, they all have skill creation skills. Right? So then there's... I have a skill cutter skill that cuts it down because they... They're all... They... It just documents every way in which those default skills that are created are bad. So you always run this after any skill is created to cut it down. Right? So... Kinda like this. Yeah. It's really annoying because sometimes there is slop and, like... But there's good in the slop. And then you you have to, like, sort of whittle down what they're... What they're... [00:43:41] S1: Right. Right. Oh, interesting. As in you're saying, like, even the slop is it's still in the language of the agents. So the slop means something. It's like, yeah. Okay. Okay. Gotcha. Got it.
00:43:41 · Mihail Eric, S2, Shawn Wang
[00:43:41] S1: Right. Right. Oh, interesting. As in you're saying, like, even the slop is it's still in the language of the agents. So the slop means something. It's like, yeah. Okay. Okay. Gotcha. Got it. [00:43:49] S2: It's fun. Everyone has their own version. Use first, there's very easy to like vibe check a skill, right? You just call it, run it a few times. And distill these preferences into a skill that's like, this is what I like, this is he doesn't like. Mhmm. And it it adjusts over time and then, you know, [00:44:05] S0: it So gets the problem is that... Yeah. Yeah. You're calling it, but the the actual... The worst part is when a skill triggers when you're not expecting it to. Like, you wrote it one time to fix the thing, and then, like, now you're working on a completely different thing, and it triggers. Why? Because it's badly written because the trigger says, like, oh, you call me whenever blah blah blah happens. Totally. And you're like, okay. Read's good. And then you do a different project and triggers and you're like, oh, what [00:44:29] S2: the hell is happening? There's like a few people that are just like, okay, I have a bunch of skills and I keep adding on sometimes. I know some people like Eugene Yan, they are pretty particular about how many skills they have. They prune them. If not used, they clean them every few weeks. [00:44:45] S1: Yeah.
00:44:45 · Mihail Eric, Shawn Wang, S2
[00:44:45] S1: Yeah. [00:44:46] S0: Yes. So, yeah, this is like a huge problem if you're sort of in that skills world. I think this is actually why I suspect there are good... There are very smart people at OpenAI and and Anthropic that want skills to die, which is not a popular statement. No. Now they they won't say this publicly, but it just feels sloppy. This feels sloppy. I I know it's... Yeah. I I can see my own skills people being sloppy. It doesn't feel AI peeled. [00:45:14] S1: Right. Right? [00:45:14] S0: Yeah. Yeah. It's it's not it's not what AGI was supposed to look like, which is... It's like everyone having a giant bunch of markdown files. [00:45:21] S2: I mean, I [00:45:21] S1: think part of it is like There's a gardening problem, yeah, in some sense. [00:45:25] S2: A lot of people actually don't look at their markdown files, right? Like the UX of it, of just chatting what you want, creating a skill in chat, using it, seeing output. There's a of value you can get without ever looking at your markdown. Don't over complicate it at some level for yourself. You can get a lot of value. If you do stuff repeatedly, just be like, look at the last five times we did this, make me a skill, and you can iterate it with chat. There's a good side of it as well, right?
00:45:25 · S2, Mihail Eric, Shawn Wang
[00:45:25] S2: A lot of people actually don't look at their markdown files, right? Like the UX of it, of just chatting what you want, creating a skill in chat, using it, seeing output. There's a of value you can get without ever looking at your markdown. Don't over complicate it at some level for yourself. You can get a lot of value. If you do stuff repeatedly, just be like, look at the last five times we did this, make me a skill, and you can iterate it with chat. There's a good side of it as well, right? [00:45:54] S1: There's another thing. There's, I guess, another way to look at it as well, because there's a, there's the use of the skill as just procedure that the agent kind of runs whenever to do a code review or whatever. But then, in the case of something like Devin, right, where we've codified skills as a way of representing our best practices, right, and then making... And exposing that to the agent, the code review system, to then, you know, institute the APIs the way that we like to define them, institute kind of the service level boundaries the way we define them. That's like an alternative way of sort of representing, like, org level knowledge, and maybe not so much this, like, arbitrary thing you want the agent to do so you don't have to write the prompt each time. And so I like that. I feel like that positioning might... That purpose might exist longer than just, you know, [00:46:40] S0: these kind of random procedures. I just Googled Monaco AI skills, and it's giving me the the the country. [00:46:48] S2: Hey, maybe the country.
00:46:48 · S2, Mihail Eric, Shawn Wang
[00:46:48] S2: Hey, maybe the country. [00:46:49] S1: One day, one day. We're gonna beat them at AEO, [00:46:52] S0: that's for sure. Can you just buy the country? Well, here's the thing. Okay. [00:46:55] S1: On this, we we, we're about to buy... No. We're not about to buy the country. But but but, you know, we we did, we dug into a little bit of flack when we, bought the monaco.com domain. I will say that. We did get sued by them. What? Yeah. There's there's like a whole public there's a whole public thing about this where... Yeah. So they didn't like that, but but here we are. And we won. We won. [00:47:17] S2: So... Okay. And we're keeping it in. I was about to say this is spicy, but you won. [00:47:22] S1: Yeah. Yeah. No. No. It's... [00:47:23] S2: Is see that why flags flying around? [00:47:27] S1: Because it's [00:47:28] S0: like a victory. It's because [00:47:29] S1: it's a victory. It's our victory. That's funny. [00:47:31] S0: For for [00:47:31] S2: those that are in SF, just look up and, you [00:47:35] S0: know, you'll be fine. So for for those who are on audio, we're we're showing this on YouTube. It's... You fly a plane around. I think you're still one of the few that actually do this on a regular basis. Sam Blond came up with it, $6,000 a day. I don't necessarily think it directly causes me to Google what Monaco is, but all of Monaco's ads are just Monaco dollar sign. [00:47:59] S2: We stand for starts money. [00:48:03] S0: The conversation, but the whole point of the ad is what the hell is Monaco?