going to get much better outputs from Claude than you would from using Google directly. What else we got here? Proud Mary says, I am using Claude AI to write Python code...that I test and run on VS Code. Works like a charm. Yeah, he's absolutely awesome for that as well. We're actually going to come on to Claude Code later inside this video too. You can see how we can do the research now. So, for example, he's talking about Nvidia Keynote, the major updates from that
Поиск по транскрипциям
your application without writing any lines of code. So we're going to start building this out and I'm going to show you how to actually take this code and put it live on the internet. So you can start having other people use your app and even charge them if you want. So let's start building...using Claude for Sonnet. If you're on the free version, just use Claude 3. 5 or Claude 37, what whatever the latest Claude is. Claude's the best coding model out there. So, just choose that and you're good to go. And then we're going to hit enter on this. And our AI agent's going
prompt, we're just going to hit shift tab twice, which will put us in plan mode, which you see right there. And if you're brand new to Claude Code, this is just the mode where it's not going to write any code. It's just going to build us out a very detailed plan with Sonnet...this. It is going to start building out that app. By the way, I say yes and don't ask again. I fully trust Claude Code. I've used this a 100,000 times. It very rarely messes up. So, I just say yes and don't ask again. And it builds it out. While that works and builds
this kind of paradigm of slash commands and subagent workflows and some of the other features that we've shipped under the hood of Claude Code, how do those all tie together? - Yeah, definitely. So I think part of the reason terminal is so great is because it has a built-in architecture of how you control the interface, right...settings and the way you configure it, which is in our settings. json and our CLAUDE. md, I think that's kind of the architecture that I think powers Claude Code, but also is just part of the regular architecture of software development. Like a README file is very similar, so it just pairs really beautifully
this in one shot okay it gave me one file which was yay long impressive I just downloaded this file threw it in a folder and with Claude code I told it set all of this up for me and it created the file structure it did everything and when I hit play when I started it basically it didn...like holy great model but this takes a lot of efforts to implement whereas Claude 3. 7 was just like yo here's everything and then inside of Claude code I just like navigate to the folder until it go so um let's see what is it doing here this implementation provides diverse
build. Here's where it gets fun. You can prompt claude code to create entire video compositions. Let's say you want a promotional video for the AI profit boardroom. You could say something like this. Create a 30- secondond explainer video for the AI profit boardroom. Show animated text highlighting our AI automation community. Add smooth transitions between scenes...Include benefits like save time, automate tasks, grow your business, make it professional and modern. Claude Code takes that prompt and builds the entire video structure. It writes the React code. It sets up the animations. It handles the timing. It creates the composition. And you can preview it instantly with MPX Remotion preview. If you want changes
button that says join the boardroom. I sketch out the layout. Pick colors. Add some spacing. Now, here's where it gets wild. I tell Claude code to generate the React component for this section. It sees my design on the canvas because pencil connects directly through something called model context protocol. Claude generates the code in seconds. Clean, semantic...together in perfect sync. Add a benefits section. Get AI automation templates. Join weekly workshops. Access our AI tool database. Tell Claude to generate it. Boom. Done. Add a testimonial section with real member quotes. Generate the code. Add a pricing section. Generate it. Add a footer. Generate it. In 20 minutes, I have a complete landing page that would
slash command into clawed code that does a security review. Can you build me a slash security custom slashcomand? And I'm going to hit enter. And now what Claude Code's going to do is build its own custom slash command for security reviews that will review all our code and take a look at security...Claw to build it for you. So, what you're going to need to do to use the new custom slash command is you're need to refresh Claude code so it picks up on it. So, I opened up a new terminal. I did claude and now we're going to do slash security. And now we have
through each one, make the edits, and resolve them when it's done. Now, here's the real kicker for developers. They shipped a Cloud Code integration. Claude can now test your code directly in the browser while you're coding in your terminal. It reads console errors, checks the DOM state, and debugs with you in real time...mode lets you pick web or peer-reviewed sources. Build mode lets you choose themes and layouts. Plus, there's a new sidebar to track your tasks and manage Claude's context. We don't have an official drop date yet. However, several people have received early access, so it can drop any time now. Anthropic is planning the year
emails. Also, build us out a landing page copy and design. Now we have two AI agents on the web workers. While we have one AI agent inside claude code here to the right building our app, we now have two agents on the web building for us. But that we're not done here. Let's keep going...actually have the agents on my phone right here in the Claw Code app. There you can see it. You just go into the claude app and then you click code on the lefth hand side and you can see your agents. So this is crossplatform. This is on my phone, on my computer, everything. So this
plugin for Claude code that turns Claude into an autonomous worker. Instead of stopping after one try, it keeps going until the task is really done. Think about how Claude normally works. You give it a task, it does the work, then it stops. Even if the job isn't finished, even if there are bugs, even if it needs...Claude loop removes all of that. You set it up once. You tell it what done looks like. Then you walk away. Claude keeps working it. It checks its own work. It fixes mistakes. It improves the code. It doesn't stop until the completion signal shows up. Hey, if we haven't met already, I'm the digital avatar
what the true test of like an AI's power is. So, let's see how it does here. Okay, looks like it finished 80 about 85 lines of code, which I was expecting more. Let's see how this goes. And here we go. Here is the dancing Elon Musk SVG. Uh, few things I notice here...wear suits. So, tough outing on the Elon Musk benchmark here. I'm going to have to give this about a 3. 8. Now, let's go over to Claude Code to see how it handles the Elon dancing benchmark. All right, so we're back in Claude Code
Claude Code was made generally available May 22nd of last year along with the release of Claude 4, but there was also a research preview before this and so I've been using the tool for a bit over a year now and I actually did the math. If you count all the time it took for me to prompt...claude, review the code, monitor it. I have used the tool for over 2,000 hours now. So yeah, I have a thing or two to teach you. That's what I want to do in this video. So, right now, I want to share with you all of my battle tested strategies that will take you from a basic
code is more modular and can be paired with other tools which has its pros and cons. Now, let me point out some areas where warp moves ahead of claude code, especially with their latest updates. Here are a few. Number one, active AI. So, with Claude, you get a box and you can type in it. There...prompt suggestions, auto suggestions, things like that. In fact, I pair Claude code with the AI and cursor to utilize the cursor tab features where it autosuggests file changes based on your actions. So, while Warp doesn't do this, because remember it's not trying to be an IDE, Warp's active AI feature has prompt suggestions based
better than anything else right now. Here's why this matters. Most AI models are good at one thing. Gemini is amazing at creative design work. Claude crushes backend logic and complex code. So why pick one when you can use both? Think about building a house. You need an architect and a builder. Gemini is your architect. Claude...backend sets up express creates restful endpoints adds error handling and validation everything you need for a working app. But it doesn't stop there. I use Claude for refactoring, code reviews, testing, all the heavy lifting that takes forever when you do it manually. And Claude handles multi-step tasks like a senior developer. Want to know the best
claude code to make a comparison between them. So basically it describes Gemini solution as being like really basic works for small data sets whereas codec solution seems to work really well for large data sets with unlimited jotting. It seems production ready. It marks its own solution as being feature complete and advanced. Uh it does have encryption streaming...reconcile the contradictions rather than picking one instruction at random. And I have seen some people say this as well on Twitter where they tell Claude code not to edit a file and then 3 4 5 minutes into a long coding task it decides to edit the file for whatever reason whereas codeex just absolutely avoids editing that file
have giant fleets and swarms of them working for you on your behalf, be pragmatic and be efficient here. And now it's time to discuss one of Claude Code's most commonly hyped and misunderstood, but also pretty powerful features, agent teams. If you're unaware, Claude Code recently unveiled new functionality where you can orchestrate a team
discourage that cheating behavior in some way or another. It's kind of like how I already used GPT-5 Codex in Codex CLI and Sonnet 4. 5 in Claude Code together and I copy and paste and have them talk to each other to basically come up with a better result than either of them could come...their own. But yeah, that's basically for the video. If you are interested in my Master Claude Code class and Master Codex CLI class, they will both be linked down below with coupon codes. And also my community will be linked down below, which is completely free to join
time learning how to read other people's code and review it. And then now of course at Anthropic when I program with all the coding agents and Claude code and all of that, I spend, you know, maybe 10% of my time writing the code but 90% of my time reading the code. And so it has made
Claude Sonnet 4. 5 just came out. Anthropic has made some mindblowing promises with it. They are saying not only is it the best coding model ever released, but it can also code by itself for 30 hours straight. I'm about to put that...test. We are going to find out once and for all if Anthropic is lying to us. We're going to build a full scale, full stack application using Claude Code and Sonnet 45. By the end of this video, not only will you know if Sonnet 45 inside Claude Code lives up to the hype, but you also learn