with the foundations. He talks about what you need to know nonI related in order to build a business. After that, we get to the fun share your screen Claude code Claude little tips and tricks and what you can do. But I encourage you to listen to the whole episode cuz I think people who are going
Поиск по транскрипциям
claude code is not at the top of that. — Um, so there's differences but I think it — it doesn't necessarily mean that the model labs are better at it. It just means that you have to understand how the models work and people who look at the at what makes a harness tick around the model...like Opus 4 5 was when they started to like really feel this. It might have also just coincided with winter break when everyone went home and started using claude code and realiz how good it was. Um but I think like around like November, December like I think there's has been this like — at least I sense
even if you're not, you can use gpt5 for free right now. Go to chatgbt. com. Open up this window, pull it to the side. So you have claude code in one window, you have chat GPT in another window. I like to use a second AI as a co-pilot for everything I'm doing...come back here in a claude code, I go into planning mode and I talk about another piece of functionality I want. Let's say I want the ability to draw in notes. So, I'm going to go in plan mode and send that. I want the ability to draw in notes. What that's going to do, obviously
Claude Code like the next step there? - Yeah, this is something I'm personally, very, very passionate about. I think terminals are like the first user interface, right? Like, they're the first ways that people used to talk to computers. They were text-only. There were like very specific commands you needed to know in order to be able...removes the need for the UI abstractions that we've had to develop previously for different applications. - Mm hm. Exactly. I also think big part of why I think Claude Code is successful is, you know, no one likes copying and pasting things, from like a web UI to like your local file. Like, I definitely do this
providers when it comes to transcription. So I'll press enter here, and then enter here. So basically reading through this quick comparison of the plans, it seems like Claude Code and Codeex CLI did do pretty well on here. Gemini CLI did do the worst because it just did not consider nearly as many things. Um, I think realistically...think I will just combine plan one and two again over here. Plan three is like basically useless can be discarded. So whilst we're waiting I think that claude code and codeex CLI are basically at the same level when it comes to planning out something. And then Gemini CLI is basically number three again over here. And there
like how is this different to openclaw or Claudebot? I'll probably do a whole video breaking that down. So if you think back to that claude code harness that I have the little ball on the cube, the cube has had millions or hundreds of millions of dollars at this point spent on it to make a really great...things you needed to do agentically. Ripping the arms off and then sticking some like for example web search. Web searching comes native in claude code. You have to add it on as an add-on to your claudebot. So that's just a little bit of a teaser if you want to get my full breakdown on like
program. Um, and then also the final thing I mentioned is um, I do have a brand new course that's actively being developed and that's for Claude Code and that is for this project. So, let me go to Figma and landscape here. It's a um let's see here. Right there. Yeah. It we're building...interface with Versal. We're going to host it on Versal. Um, to interface also with Stripe. We're going to integrate Stripe billing the billing for and have Claude code be able to control the billing aspect through MCP. Um, we're also using Chrome DevTools MCP. I'm trying to remember this stuff
video, I'm confident that you will be able to do the same thing regardless of your technical experience. And I did all of this with the Gentic software claude code that runs inside of a terminal window on my Mac. Now, if you find terminal scary, fair enough. But this tutorial is designed to lead you through this step...listens to your commands in natural language. just like a developer you hire would. Let me show you how. This is free software from Enthropic and it's called Claude Code. It runs inside of your terminal which comes pre-installed on every Mac and Linux machine. You do not need a subscription just to use the service, although
premium domain? Oh, Vibe Assembly. AI is $50. I don't know if I like Vibe Assembly, though. Vibe Assembly. AI Vibe Assembly. Have you faced a situation where claude code opus45 tried to cheat instead of fixing a failing test during appdev but adding any type the laziness is gone? Uh I haven't faced that with opus...worry about that some other time. All right. We'll come back to that. We'll noodle on that for a little bit. Let's go back to Claude Code. What else do we need to ship before we go live? Finship
right now I'm in Visual Studio. I have the Claude Code extension opened up and installed. Yesterday we covered all of Sonnet 45 in the big release. Anthropic has made some huge promises. Let's see how well they do here. We're going to be testing out this Claude Code extension as well. This is something they newly
detect it. Claude code crashed IBM stock 13% in a single day. Worst drop in 25 years, $31 billion gone. Why? Cobalt, a programming language from 1959 that still runs most of the world's banks and airlines. Modernizing it the traditional way took years, cost a fortune. Anthropic published a post saying Clawude code can do the same analysis...enough. If you work anywhere near legacy enterprise software, finance, banking, insurance, the modernization backlog that's kept consulting firms alive for decades just got a serious competitor. Claude Code
error. This is actually a good thing. This is a good thing. I'm going to show you how to debug. Debugging very, very important with claude code. It's like the number one question I get is what happens when I run the bugs. I'll tell you exactly how to do it here. When you're looking...going to look at the error. It's going to figure out what's going on in 99. 9% of the time where I just give an error to claude code. It figures out how to do it in one shot. So, it is going to start working here and is going to fix our error
results you can get out of this. Right after I'm done showing you this demo, stick around. I'll actually give you the prompt you can put into Claude Code to build out this demo so that you can have your own deep research Kimmy K2 thinking app that you can use your own AI agent. I built this...with one single prompt. This was one shot with Claude Code and you're going to see just how powerful this model is. So, with this demo tool I built, which you can build out after this, you're able to enter in any sort of research prompt you want, and it will use Kimmy K2 thinking to do this
going to hazard a guess here. guess. Trading with Will. I'm going to hazard a guess that 80% of your subscriptions are on things significantly less important than Claude Code. I'm going to hazard a guess that 80% of your subscriptions are on things that if you got rid of them today, your life does not material change...sort of way. If you were to delete my Claude code subscription right now, my life would materially change for the worse. I would be contributing significantly less economic value to this world. My guess trading with Will is 80% of your subscriptions you have today, you can delete them right now and no one you would never notice. Open
commit PR which basically does exactly that. it creates a new branch, does a commit, and then makes a PR into main. But as of a couple months ago, Claude Code began to treat slash commands as skills, which means that if you do like /context, then you will see that as a skill instead. But of course, with skills...like this. And then opening it up, I can add front matter to the very top, kind of like this, including the disable model invocation, which means that claude code will never trigger this itself. only I will be able to trigger this. So these ones I basically repeat across many different projects and my user level and then
worry, I'll do that soon enough. Right now, I'm actually holding a community event later today, showing the community some of the tricks that I learned in Claude Code and some of the things that you can create with these parallelized agents. But yeah, I just wanted to start to sneak that into this new show because...people are bastardizing it already for writing life management and all other types of use cases. If you're curious about that, I would recommend you check out this claude code page on X. They share different projects and releases from the community and Enthropic. They're not affiliated with Enthropic, which at this point, I also want to point
perfect entry spot. I would say start with lovable cuz you can do like you can do most of this in lovable but this BMAD method and claude code is like it's level two. Lovable is level one. Level two is to go through once you're done with lovable. There's certain hiccups there. Level...sure that there's no security risks. all those kind of things. And then yeah, that that's that would be it. Step one, learn lovable. Step two, learn Claude code and BMAD. Step three, build your own product practice. Build a templated approach to solving a particular problem for a niche. Step four, build a demo for that
thing that you want to build. — Yeah, the planning step is is massive. As I've got deep prints like aentic engineering, the plan is everything. And Claude code is so good at it because it can read every document you provided it to provided it with and then it's writing that really comprehensive spec sheet breaking down each...look back on and execute. So asking for plans rather than I think that's the difference between like a complete new beginner or like user of claude code and like the first step is realizing that the value is in the plan. So the spec or the brief is a is everything. So let me uh give
Claude Code will summarize a conversation, kind of like the /compact command up until that point and then start a brand new conversation. So just pressing enter, it will compact it and then restart a brand new chat. The ask user question tool in Claude Code can now display markdown
competing directly with paid commercial models. But here's what really matters. It's about practical use. This model works across a ton of different tools. Kilo Code, Claude Code, Rode, Droid, Blackbox AI. You can download the open weights from hugging face and run it locally. That means total control. No cloud dependency. You own the whole stack