00:16:11 · Thariq Shihipar, Aditya Agarwal
[00:16:11] S0: Yeah. This is really, really hard. I I think I think that the art of like this, like, harness engineering, which is... Can be the harness, it can be skills, it can be how you prompt, is like a really rare skill right now. And that, like, it's hard to know if you're even good at it or not sometimes, you know, because... [00:16:29] S1: Is it still like... Is there still... Is this still mostly human taste even with Anthropic? Or are you starting to do a lot of this recursively where you would try to imbue meta agents to help kinda get everybody to do this properly? [00:16:44] S0: Yeah. Well, so I mean, Anthropic is not like a single, like, you know, monolithic entity, and so everyone has different opinions. Right? And I think that this, like, self recursive, like, harness thing is... Some people are into it. I think other people are not. You know what I mean? And I think that, like, it's Of the Claude Code. It's a mix. Okay. Yeah. Yeah. I I think we try and see what works. But I I think like something like workflows, for example, which I think is huge and I think I'm excited for everyone to try it out, is definitely something where we saw it internally just do really well and we're like, oh, shit, this is just sick. And it did help drive some results like the Rusty Wright. But yeah, also we like, you know, we're... We use the models every day and and it was better.
00:16:44 · Thariq Shihipar, Aditya Agarwal
[00:16:44] S0: Yeah. Well, so I mean, Anthropic is not like a single, like, you know, monolithic entity, and so everyone has different opinions. Right? And I think that this, like, self recursive, like, harness thing is... Some people are into it. I think other people are not. You know what I mean? And I think that, like, it's Of the Claude Code. It's a mix. Okay. Yeah. Yeah. I I think we try and see what works. But I I think like something like workflows, for example, which I think is huge and I think I'm excited for everyone to try it out, is definitely something where we saw it internally just do really well and we're like, oh, shit, this is just sick. And it did help drive some results like the Rusty Wright. But yeah, also we like, you know, we're... We use the models every day and and it was better. [00:17:36] S1: So... Switching gears a little bit. And by the way, as I mentioned, if you have a burning question already, just raise your hand and we'll kinda get you in and I'll continue to ask. I see Jason there, so we'll have you queue up. One quick one while Richard's listening to that is like when you came to startup school, you shared some of your favorite prompts. Like they were kind of meta prompts. Right? The the types of things that you can ask the model to make you better at your building. Can you share some of those again? Because I thought it was helpful or some of the the new ones. And I forget specifically this... The specific one you shared at startup school, but I I I love that line of thinking of just like, are the meta prompts that that are useful?
00:17:36 · Aditya Agarwal, Thariq Shihipar
[00:17:36] S1: So... Switching gears a little bit. And by the way, as I mentioned, if you have a burning question already, just raise your hand and we'll kinda get you in and I'll continue to ask. I see Jason there, so we'll have you queue up. One quick one while Richard's listening to that is like when you came to startup school, you shared some of your favorite prompts. Like they were kind of meta prompts. Right? The the types of things that you can ask the model to make you better at your building. Can you share some of those again? Because I thought it was helpful or some of the the new ones. And I forget specifically this... The specific one you shared at startup school, but I I I love that line of thinking of just like, are the meta prompts that that are useful? [00:18:15] S0: Yeah. I... I'm trying to work on a post about this as well. I I think that like [00:18:20] S1: Follow Thariq on Twitter. He's already... What are you? Like, 250,000 followers now? It's a big deal.
00:18:25 · Thariq Shihipar
[00:18:25] S0: Yeah. I I think that like the way I think about it is, okay, when you have a task, you have like no knowns. Like like you know you want this and... But like maybe is it in the prompt or not? Like, did you write it down? You know? Then you have unknown knowns. You know? So you're like, you you think you want it, but like you don't know it until you see it. You know what I mean? And then you have like unknown unknowns, where you just like don't know that this is possible, you know? And you like don't understand the system well enough or something. Right? And then you have like known unknowns where you're like, oh, I I don't know how this works, you know, but I would like to figure it out. Right? And and so I I think for every problem, analyzing it along those dimensions helps me pull out like, okay, this is how I should prompt the model. Right? And so for example, like if I'm just being lazy about prompting, sometimes what I like to ask the model is like, hey, like interview me about this problem. You know? So like here is my idea. Let me like interview me using the ask user question tool, and then I'll answer. So like that's one example. Sometimes if it's like, okay, I'm working on a design and I know what I... I'll I'll know it when I see it, you know? I'll ask it to like do a bunch of explorations. And so I'll be like, explore this design and make a bunch of mock ups in HTML and show it to me. And then of those like whatever four, then I might ask it to do more like exploration there. But like, again, it's like in my head, I like I have some preference, but I I don't know yet. Right? And I think a meta prompting technique there is like, if you're still figuring something out, do you want to do it in a cheap way? So like, you know, if you're doing UI, it's cheaper to do it in HTML because it's fake. You know, like you're not rewriting the state and the like, you know, all this other wiring up stuff. Right? Like if you ask a model to add a button, in HTML it's just a button. But inside of your app, maybe there's a new route. There's a new like reducer, all of this stuff. Right? And so in the same way that like, you can do this for like back end endpoints and testing and things like that too, just like, what's the cheapest way to figure out like, is this the thing that you want? Right? So... Yeah. Jason?
00:20:52 · Jason, Thariq Shihipar, Aditya Agarwal
[00:20:52] S2: Do you have any advice around whether we should say thank you? Because it it does... I'm imagining it probably is a good signal of you did good, you know. It's like, yay. Or is it like just a complete absolute waste of tokens? [00:21:06] S0: This is actually like an interpretability question, which I like, you know, I I learned a lot about a Goodfire. I think that like probably... It it's hard to say. I don't think we run like a thank you eval, you know. But we have done some research recently about like emotions in Claude. And like, you know, if you are mean to it, it does activate this like, person is being mean to it feature, you know. And it does guide results in some way. I I think that like... I don't think we've done like swee bench against like meanness or something, you know? But yeah, I am generally like positive for the models. Think in the same way that you should be like, you know, nice to most things, you know. And and the very least, so that you're not embarrassed to share your transcript with me. Because I definitely get run into bugs sometimes. I'm like, oh, great. Like, I can help you just like send me the transcript. And they're like... That reveals you're a terrible person. Yeah. [00:22:06] S1: But I've literally had people be like, please don't judge [00:22:08] S0: me for this. It was like a really long and tough night. I'm like, okay. I I got it. I got it.
00:22:08 · Thariq Shihipar, S3
[00:22:08] S0: me for this. It was like a really long and tough night. I'm like, okay. I I got it. I got it. [00:22:13] S3: Yeah. I have a quick question about how you think about the design of Claude Code and the harnesses. One of the things that is kind of emerging across the landscape is that we're seeing convergence across things like different skills or approaches. And I'm wondering internally if there is any type of debate around that, whether they're gonna be different architectural approaches across the different platforms or whether you believe that there's gonna be one unified way to do maybe a couple things or even everything. [00:22:41] S0: Like, are you asking across models? Like, if the same skills across like Opus and GPT or like just Well, things like memory, learning, question planning, they they were a [00:22:55] S3: little staggered in terms of rollout in different different platforms and harnesses picked it up. But the thinking behind that is is probably quite similar in terms of what needs to be true to make something work.