--- title: "Thariq Shihipar, Independent" speakers: - name: Thariq Shihipar affiliation: Independent room: "Continental 4–6 (Mainstage)" starts_at: 2026-11-12T17:30:00-08:00 ends_at: 2026-11-12T17:50:00-08:00 format: talk recording: 2026-11-12_1724_Thariq duration: 1173.3 source: qwen3-asr-1.7b --- > The Mainstage ran ahead of the grid this afternoon, so the recording starts before the scheduled slot. > > The MC's introduction is captured at the start. The program lists this session only under the > speaker's name; the operator's notes give the talk's title as "The Sweet Side of the Bitter Lesson", > which the program does not carry, so the title above stays as scheduled. > > Transcribed locally with Qwen3-ASR 1.7B (timings per ~5 s segment), prompted with the > speaker names and affiliations of overlapping sessions. "Thariq Shihipar" at 00:00:09 and "Dexter" at 00:12:44 may > come from that prompt: the unprompted Voice Memos transcript has "Zarik Shinbar" and "Dex" there. > "Anthropic" was not in it. Other names are kept as transcribed: the speaker calls himself "Farward", the > title is "the sweet side of the video", and Claude Code is "Cloud code". The Voice Memos transcript is kept beside it as `transcript.txt`. ### 00:00:00 — Introduction How we interact with AI-powered tools. Ladies and gentlemen, please welcome independent engineer and former Anthropic team member. Thariq Shihipar. But so it's not World of Warcraft. I'm not sure what I'm doing here. But hey guys, I'm Farward. Nice to meet you. Yeah, I wanted to talk to you about a couple things I've been thinking about since I left Anthropic. It's been a wild year, and there are a bunch of reflections I've had. And my conclusion is about. One calling the sweet side of the video. And the ways I've been wrong. So I think I got a lot of things right. Ah. I think I was right about models being able to edit and generate videos. I was talking about this like. Sometime in like June last year. I was right about models being able to use HTML to understand their work. And right about models being able to help you. Understand and figure out your unknowns to help, like plan. And learn about things. So these were all things I was like right about, and I was happy, and I think that like. A lot of you might be incorporating sort of stuff in your work. ### 00:01:40 — What I was wrong about But I've been wrong a lot too, and I think this is one of those things in AI where you sort of just like. Memory hold the things that you're wrong about, and you're like, oh, like I was just always expecting, you know. Yeah, to be great, but here are some of the things I've been wrong about. So. First, like Bash is all you need. I think there was this like moment where the agents were really good at using Bash and. Cloud code was built around the Bash tool, and so you know people are like, well, we'll just turn everything into a CLI, like you know, like. There is just going to be bash and CLI's, and that's going to like take over the world. That's actually kind of not how it turned out. Like I actually prefer MCP to CLI's now. You know, but now there's also code mode and. Skills and disclosure. Like there are so many other like ways that agents can you know. Act on the world, right? And the fact that they were using Bash was just like this weird remnant of you know like. Them being trained in a certain way at the start. Then there's like React is all you need. I think. For a moment, it seemed like the models were really, really good at running React. And so, you know, if you wanted to create an AI-enabled app, you should figure out how to use them. You should do React Native. You should, you know, like you should just make everything on the web. And that also has not been true. I think, like now, you know, models. They can write in HTML. They can write in React. They can write in Swift. They can write in the Figma proprietary JSON format. And they'll all look roughly the same, right? And so. I was afraid that like you know the models had been trained on React and so we'd just be writing React forever and React would be like the last language almost right and. That's almost certainly not true, right? I think a lot of people are running their own engines now. And I think one of the other things I was really concerned about was like diffusion models will kind of like eat art, like and just be like. You know we wouldn't have any control over these models, and or the kind of art that these models are creating. But it would just like create the most beautiful stuff, and that hasn't been true at all. Like I think now LLMs can make art through. Code through controlling your computer and using you know Blender or Photoshop or. Yeah, finger or whatever you'd like, right? And so. These were all sort of like fears I had about like I think. The bitter lesson, you know, our alien intelligence. The show guys, right? Like, like these are all words we use. I think to. Describe this kind of like fear that the models will just have this like way of doing things, and it will be weird. And alien to us, and we just need to accept it. You know, like we just have to like submit to this like foreign intelligence. And that's just like I don't think that's what happened, you know. ### 00:04:28 — AGI-pilled and human-centered And so I think the like the question I come back to a lot is like okay like how can we be AGI built and human centered. And what can we learn about you know this progression? What we like. You know what can I learn about what I was wrong about, right? So I think it's really important because software engineers are like the leading edge of like AI adoption, right? Like. Anything that is running the software will happen to New York and other parts of the economy, and so. I think how we deal and engage with AI is, yeah, how other people will as well, and. You know I really enjoyed the different talks here, talking about yeah fighting slop. I think you know we're the first defense. For that, right, or like being more ambitious. ### 00:05:09 — Industrial revolution and Renaissance What I like to think about it as AI is both at the same time an industrial revolution. And the Renaissance, right? And I think this year was really about. The industrial revolution, right? Like we use the word software factories very often. And like just sort of like turning software from its like. Craft that was very unreliable and hard to execute on into something that like. Is not the limiting factor, right? And this sort of like industrialization is absolutely happening. This is definitely. Something that is happening with software, but at the same time, there's also kind of a Renaissance happening, right? Where people want and higher and higher quality software, and every marginal. Bit of like creativity, utility, like spirit. Like soul that you put into it is like is useful and like economically valuable, right? And so, simultaneously, got both of these things that are very hard to juggle, and sort of like. Hold in your head at the same time, but I think it is very true, and. Yeah, we're doing both of them. ### 00:06:19 — The jagged frontier and the smooth interior So, oftentimes when we talk about AI capabilities, we talk about like the jagged frontier. Right, and so, you know, maybe like you go back two years, and you're like, asking AI to like write this function for you. Or you know, like a year and a half ago, a year ago, it's like build this feature. And you're like, hey, build this like to do app of like a kind of simple to do app, and then clone Slack for me now, right? And. And now it's like if you build Mario Kart from scratch, you know, and then like port it to like the Wii or something. There's like. The capability increases are the things that you see, right? Where, whenever you look at the Evals and benchmarks and stuff. You see this like just. This giant frontier of our capabilities, but I also like to talk about. Like the smooth interior. Where the way the way the models. Is there an animation here? The smooth interior is like the idea is that the models are. As they get better, the animation here is like this jagged frontier and this circle in between. And so, the circle sort of represents all of the things that the models are too are smart enough to do. And so, like examples being like writing React code, right, or like using Bash. And what you see is that like it's not jagged below the frontier, right? Like if. It can write React. It can write Figma. It can write JSON. It can do sort of like all of these capabilities all together. And there's like a breadth of tasks that the models are really good at. And this is something that's also, I think, really like important and. For me, even more impactful because I don't really care that. They can build Mario Kart from scratch anymore, right? What I really care about is like. Hey, is the software that we're building is it quality worthy of like? A renaissance, you know what I mean? So. Yeah. Like I think that like you can do things the way you want. So like if you're building mockup, you could be like, hey. Can I get an HTML? Can I get a React? Can I get a Figma file? Can I computer use my computer and use After Effects? Make a video. There are all these different ways of getting the like final result. And I've been really happy with this. I really was afraid that there would just be like one thing that the models loved doing, and we'd only be doing that. And you know, like taking to the limit, kind of like you know we talk about this a lot, but like. You know everything we've talked about so far about like you know how you build like these harnesses and software factories and. These like the scaffolding for the models, even things like you know for a while we were also like. Hey, the model you'll like. Your job is to QA the model, you know, or review the code. And that's clearly not the case, right? I think as models get smarter and smarter. It will become more of this, like smooth interior, right? Where they're just doing things well, right? And like the perfect colleague can imagine won't need to like. Really verified its work or scanned for security vulnerabilities. We'll just like write the code and be like, yeah, I'm done now. And you will check and you're like, oh, you're right, like. You know, according to the spec and the goals that you that we did, you wrote you did it correctly. And I think that's actually very exciting. ### 00:09:42 — A map of what the model knows One of the ways I'd like to sort of think about. Visualizing how much the model knows of just eval. I think Caparthy posted it. But basically, you ask any these are Claude models. You ask them. You give a longitude, latitude, and you say like, hey, is this point? Earth or water, right? Ground, sorry, land or water, and it. From that, you plot a map of deeper. Right, and yeah, this is Opus, 4.6 is. Somewhere over there, right? We've got a good software engineer model. Kind of like a blurry view, and now you look at Opus. 5.5, and it's like almost perfect, right? Like it knows every piece of you know. These like these coordinates. It knows like the entire world, basically. And so this, I think, just gives you an idea besides like you know the benchmarks about like the. The jagged frontier. I feel like that's less and less important to us, and more. Important, just like the general knowledge that the models have. And what's exciting, I think, about this is that we can use. Intelligence to understand and shape how we use intelligence, right? And I think that would be like the theme. Of the next year, right? Where this year I think was really about the industrialization of software and. And creating this like reliable way to make software. But what does that look like? ### 00:11:06 — Building a game So I have been building a game. It's this like thing I've been posting on Twitter about. It's a 2D like isometric game with like. Kind of like a fighting game inspired by things like Avatar and Smash Brothers, and. I have kind of like it's fun, but it's also quite frustrating to do. Like I'm really trying to like. You know, like be like a great bod coder here. I don't really know what each character does sometimes because like. I will say like, hey, I want this character to throw a rock, and then I'll be like, oh yeah, great, but like, there are all these like different. Things that can happen when you're throwing a rock, like this character can also summon a wall. So this the rock. Go through the wall or not, and like how, what speed, or like what, what about size and charge and all these questions. And I find that like the ball just makes a bunch of choices for me, and then I try it out. I'm like, wow, I have no idea what that actually does. I get distracted pretty easily. I don't feel in the zone very much. You know, I sort of send off a prompt and they start doing something else and I come back. And I honestly make less progress because of these things, right? And so. I think this is frustrating, and I think for a while I was like, "Man, I guess this is just like how you work with AI. Is it just like what it's going to be like?" You know, but I think reflecting, I'm like, man, I don't think that's right. I think that like you know, we have this like. The AI can work in the way that we want. Yeah, I don't think it's acceptable. I don't think that this is what we used to run. Right. I don't think this is what leads to like incredible software. And part of it, you know. Going back to what Dexter said, like great work is just about like this messy process, right? Of like you don't know. Like exactly what you want. It's more about the choices you made and what wasn't involved. Or was it what wasn't included versus like what was included, and so. Yeah, I think there's like. You like kind of just have to be an expert. You have to like really know your stuff in order to make something great, right? In order to like. Make something worthy of a renaissance, right? And I just like. When I look at something I'm God putting, I'm like, dude, I had no idea. Like I just don't really get it, you know, and so. What am I doing? How am I fixing this? ### 00:13:27 — A human-readable spec One of the ways I'm doing, and I'm still quite early on. I'm not saying that I've like figured it all out. But I think one of the ways I'm doing it is that I have two parts of my codebase. I've got like this like engine part of the codebase, and then I have a folder. Like I think I call it spec, but I think I'm just calling it human or something. And that folder is the code that I write. You know, and it's like this code that's lintered. And like, there's all these rules for Claude about like what that code looks like. It's supposed to feel like English. It's kind of like a spec, and I'll show you what it looks like. But it's at the density where, if I just read this code, I know what's happening, and I can edit it. I can tweak it. But more importantly, I just like you know can have a bit of a ground truth of like what the what my game looks like. ### 00:14:15 — Demo And so let's do like a quick demo here. By demo, I guess I mean I'm just going to. Whoops. Okay. Here we go. Okay, so. This is my game. I've got in here this folder game, and then I've got here. This code. This is the like agent code, right? Like this is the code where I'm like, dude, I don't know how. You know cameras work or whatever. You know, like there's a bunch of code here that I don't understand. But then, in this code, I have all my characters and I have their abilities. And so I can go in here, and I can say like, okay, this is the rock move, right? And inside of this move, there's a bunch of ways to tune it. It's well commented. It uses things. Like meters, you know what I mean? Instead of like X, like point five or whatever. It tries to like create a bunch of different abstractions along the way. And so I can see that like one of the things that you know is useful is that like there are three ways to. Throw for this character to throw a rock, right? Like it can grab it from the ground, it can grab it from the wall, or something. And I can see sort of like the. The true logic of the character, right? I don't really know. How the engine works, or how you know, like this was implemented underneath. But I've essentially created this like almost a custom scripting language for my data. And so that means, like you know, when I play this game. I can tweak little things, you know. I'm still I'm still like working on it, so like I can say. I want the rock to be a lot bigger, so I can say like. No 1.5, but I'm using an actual IDE, and so if I go here to my game. You know, I can see that like, I have a rocket and charge it. It's really big, and I throw it right. But I'm like kind of. It's really interesting because I'm also. Starting to use cursor more, and I'm using the tab model more. You know, because that now I can be like. Oh hey, what if I just want to say like, you know? If you summon a wall in the middle, like queue it up as an action, and then cursor will sort of like. You know, do the tab completion. And this is kind of like it's kind of like spec-driven development, you know, but the spec is like. This, this scripting language, basically. And yeah, it's like. Is this the final way I'm going to write this game? I'm not sure. It's like you know, in some ways. When I need to do something complicated, I still need to have Claude to do it, right? And I need to have Claude to like make that primitive. But it's actually really easy for me to understand what's happening. And I feel like in control. And while like pod is working, I can also sort of review it and work with it. ### 00:17:04 — A Renaissance of software And that's been really exciting, and so I think. The takeaway for me has been that like. The takeaway for me has been that like I now have to think about the way. The best way for me to deliver high-quality software, right? And I'm not as constrained and limited by like the model's capabilities and like what it likes to write code in or something. I'm not thinking about like. I need to think about like. How can I deliver great software? And so. I think if we like really internalize that, I think that this the next year will be about a renaissance of software. And you know, I think many people have talked. Today and you know this conference about how to be more ambitious. I don't think anything any of this is quite new, but I just think there's like. Many things to do, like new languages, right? Like. Obviously, you know we want things like Rust that are very efficient, but we also want things. To be potentially modular and like hotly loadable, we want it to be potentially readable. Right, new primitives like that. There are probably new types of databases, new types of servers, like. New operating system interactions. I've been sort of like vibe coding a bunch of personal software. That sort of messes with my operating system. Their new interaction paradigm is possible. A lot of like old HCI research is now like very interesting to revisit. New aesthetics. I think the way that we like use. And design software can change quite a lot. And yeah, I think that like there is so much to do here that I think I don't think that any of this is. Everyone, every speaker, I think, has talked about this. And I don't have too much more to add. I think that like what I really wanted us to say is that when we. Need to do is to decide what the future of work looks like, right? And decide what our ideal experience is. Because the models are not, you know, this alien intelligence that's like trying to force us into a specific way of working. They can kind of do everything, and they will, but our job is to like bring about the renaissance. So. Thank you, guys. It's kind of nice. And now, before Code Summit.