A few days back, we had made a video on Zai, which is a revolutionary new open-source AI code editor. It's completely free, and it gives users access to Anthropic's Claude 3.5 Sonic API for free, which is quite unique as they have actually partnered with Anthropic for this. It's truly a great advancement for the AI space as it introduces a new way for developers to use AI directly into a new editor with an extendable text-centric approach. Now, I recently spotlighted this breakthrough on Twitter, but in just 7 days, there's already another free open-source AI code editor called Meltly. Meltly is designed to understand your work across everything from the terminal to GitHub, and it's going to collaborate with you to write production-ready code. It's claimed to boost productivity for engineers and developers by 10 times, which is just simply incredible. Now, check out this demo video, which shows a prompt being sent into Meltly asking to refactor a file into UI components. The AI is

processing the request, and it's going to deploy agents to create a plan, and then asks for approval before executing the test, like creating components using plugins, as well as apps to carry out the command, and finally, you're going to receive the refractor code ready to accept and deploy after a couple of seconds when the agents are then deploying it. And in this case, you can even publish it as a pull request on GitHub, which is just absolutely insane. So, it's going to start committing the changes, and there you go, you have the code on the right-hand side, and you can then accept all the changes that have been made. Now, take a look at this demo video where the creator of Melt Le is creating a web app with Melt Le. Okay, it seems like lots of people want to try out Melt Le, so I'm going to use Melt Le to make a landing page to collect the sign-ups. So, let's say, can you make the sign-up button link to this Google form I made? Where's the form? Okay, here is the index page there.

you go okay now it's writing some code. Okay, let's try it out. So now when you click hack the system, nice goes to uh. Let's also embed my tweet so people can see a demo video. Uh, can we also embed my tweet so people can see a demo of how it works? Here's the link. Okay, let writing some code. It's adding some styles. Okay, says it might not appear immediately in developer mode. Oh, there it is. Now before we get started, I definitely recommend that you take a look at the Patreon page so that you can access the new subscriptions that will be releasing this week. Now, wasn't that amazing? This was something that was developed by Charlie and Jackson. They're basically friends and founders of the AI code editor. They basically developed this because of the limitations that was occurring with many existing AI coding tools. So they decided to create Melty, the first AI code editor, and it's going to understand what you're doing from the terminal to the GitHub repositories and going to help you collaborate, produce production-ready.

code now. This has been in development for only 28 days, which is kind of nuts to me, and it has basically used Melt Le to already write about half of its own code, similar to what AER does basically. Now, unfortunately, you can only get access to this by filling out a form, and if you are picked, you're going to be able to gain access to MLY in their beta access. But you can actually access the source code at the moment, which might not be helpful for many of us, but you should definitely get access by filling out this form, which I'll leave a link to in the description below. Now, one of the great features that stand out to me is the ability to navigate large code bases with Melt Le. It's going to easily navigate through large and complex code bases, and it's going to do this by understanding and adapting to your codebase. It acts like a pair programmer, similar to what AER is, and it's going to keep track of every change you make. Now, this feature is going to allow you to

quickly find and understand different parts of your project is going to make it super easy for you to manage and work with extensive code. Another feature is that Melt Le can automatically write its own commits, meaning that it can track the changes you make and generate commit messages that can accurately reflect what has been completed. This integration is with tools like GitHub, and it's going to help streamline the coding process so that it ensures that your version controls are always up-to-date, like information without needing manual inputs for each commit that has been created. In essence, Mly is basically developed to help you understand your code better, not worse. It's going to watch every change you make like a pair programmer. It's going to learn and adapt to your code base and integrate with your compiler, terminal, and debugger, as well as tools like Linear and GitHub. Now, I want you to take a look at another demo video which will showcase more about Mly. Here's a quick demo of an AI pair programmer I've been working on. Uh, I'm not going to say much more about it.

let's just get into it. Uh, let's make a new folder called animal farm, and can you put in three animals as Python files? Uh, so the first thing that this app can do is write files and do it all at the same time. So, here we have, uh, looks like a pig, a chicken, and a cow that are all getting put into an animal farm, and then on the right here, we can see a get diff for it. Nice. Um, second thing this thing can do, uh, its code name is Melty, is refactor. Um, so let's, uh, uh, make a new animal class and have all these animals inherit it. Um, Melty can also handle multiple requests at the same time, so I could also say, Also add a draw function to all the animals that make ASCII art. Uh, finally, give me some kind of interactive command line interface I can play with the animals in. So, I'm going to kick that off, and uh, what it's doing now is actually kind of complicated because it has to refactor

so it has to understand the structure of the files it's created. Um, then has to make edits in all those files at once. Um, it also has to add the draw function, so it has to keep that context separate from the other refactoring. And then it has to write the terminal command I can call. Um, so it looks like it's finished. Um, and it looks like I can run it by executing the farm.py file. So navigate to the animal farm directory. So I'll do that now. CD Animal Farm. Uh, did I spell it wrong? See. Oh. Animal uncore Farm. Uh, this is all the git diffs that it's giving me. And then I run python farm. So let's try that. Python farm.py. Welcome to the Animal Farm. What would you like to do? Uh, let's uh, list the animals, I guess. Okay, we have cow Bessie, chicken Henrietta, pig Porky. Uh, what else can we do? Let's hear an animal speak. Uh, let's say Porky. Porky says oink oink. Now let's draw.

the animal. So, let's draw Porky, which I think is three. Okay, that doesn't really look like a pig, but uh, close enough, I guess. AI can't do everything yet. Uh, so that's a quick demo of something I've been hacking on. Um, if you'd like to try it, let me know, and I can uh, uh, send you a link. Thanks for watching. Now, people have been already creating a lot of things by gaining access to MLE. In just 20 minutes, someone created a custom stop-motion app that was perfect for him. Now, in this case, this is something that MLE was capable of creating, and it turned the idea into reality, where it first off started by creating a simple decent app that was fully coded out by the AI that MLE had deployed with its agents. Another very simple app that was created is an emoji translator. Now, I know this is not some crazy web app or some super like really nifty UI component that was outputted, but you can see that the functionality of the app is there, and it

was fully created with the help of mly, and in this case, it translates text to emojis and emojis to text. I definitely recommend you guys to gain access to this by requesting Early Access so that you can get started with this right away. I have huge hopes for whatever they develop 'cause this is definitely a great step forward for the open-source field in the AI space. So, I'll leave all the links that I use in today's video in the description below. Make sure you take a look at my Patreon page so that you can gain access to different subscriptions completely for free. Make sure you follow me on Twitter, a great way for you to gain a lot of different news about whatever is happening in the AI space. And lastly, make sure you guys subscribe, turn on the notification bell, like this video, and check out our previous videos so you can stay up to date with the latest AI news. But with that thought, guys, have an amazing day. It's for positivity, and I'll see you guys fairly shortly. Peace out, fellas.

Melty: Opensource AI Code Editor - Create Apps EASILY For FREE!

World of Ai is here to push the creative use of Artificial Intelligence (AI) applications for simple day-to-day tasks!

Here at WOA, you will find tips, tricks, and guides using AI applications to help you out in your day-to-day life. Going to strive to make your life easier! I will show you how and what you need to do to help your life out! I will strive to show you the wonders of AI and if that interests you, then make sure you subscribe to the channel now!
