Transcription
This is a video I've been meaning to make for a long time, and a lot of you guys have asked for it. Today, we're diving into Augment Code, the first AI coding assistant built specifically for software engineers working with large code bases.
Augment Code is a developer AI platform that helps you understand code, debug issues, and ship faster by having full contextual awareness of your entire code base. Unlike other AI coding agents like Cursor, Pine, and Wind Surf, Augment offers a powerful free alternative with unmatched speed and intelligence.
What really sets Augment apart is its ability to fully comprehend large code bases. This is a challenge for many AI coding tools; Copilot and Cursor don't offer the same deep understanding or speed that Augment brings to the table. And the best part is, it's partially free: you get access to 3,000 messages per month at no cost, plus you get unlimited completions, instructions, and so much more.
You have the ability to instantly chat with Augment to ask and have it answer questions for you about the entire codebase. You can use the next-edit feature that gives you codebase-wide suggestions based on your most recent edit. You get auto-completions with real-time suggestions that know your dependencies and style. Another feature that I haven't really seen anywhere is the ability to use Augment in Slack. With Augment in Slack, your team has an always-available Slackbot that answers questions even when your team is offline. You can see, within this demo right here, how you can use Augment to chat with your codebase within Slack. You can use Augment for easy onboarding, making it easy for new joiners to ask questions like, "Where do you see the data fetching in React?" Augment in Slack is going to essentially answer these questions for you quite easily.
But aside from all of those features, Augment Code is going beyond just code completions by fully understanding large code bases, delivering fast responses, capturing team knowledge, and aiding in complex system architecture understanding. And what's great is that they never train on your proprietary data, so that's a great add-on.
With all of this out of the way, what we're going to do throughout today's video is explore Augment Code further in detail by showcasing what you can truly do with it.
Before we get started, I just want to mention that you should definitely go ahead and subscribe to the World of AI newsletter. I'm constantly posting different newsletters on a weekly basis, so this is where you can easily get up-to-date knowledge about what is happening in the AI space. So definitely go ahead and subscribe, as this is completely free. To get started, you can install this for VS Code or JetBrains, and you can install this as an extension. So go ahead and make sure that you have your IDE installed, and then you can go over to something like the Visual Studio Marketplace if you're going to be installing for VS Code, and click on "Install," and then have it open up within VS Code. Once VS Code is opened up, you can then go ahead and click on "Install" to install this extension, and then you can then access it on the left-hand panel where you'll see the Augment icon. This is the chat interface of Augment, where you will need to first sign in so that you can start working with it.
Once you sign up or log in with Augment, you're going to be greeted with this dashboard. This is where you can start utilizing all of these four amazing features of Augment. But first things first, you want to start off by opening up a workspace. So I went along, and I have given it access to my Deep Seek web crawler, which I—is a project that I've actually created, and I'm going to go over to Augment, and I'm going to enable indexing. Once it gets full conceptual awareness of my codebase or my workspace, we can then start working with it. You can see Augment instantly was able to sync with my codebase, and it was able to provide me a summary. Now, if you actually read through what it wrote, it was able to accurately understand each key component of all the files and folders I have within my directory. So if you actually take a look at this, you can see that it explains what this is—which is a Deep Seek WebCrawler, a web scraping framework that combines traditional web crawling with AI-powered data extraction from Grok's LM API. It was able to understand the key features where I have it so that it could be outputted in CSV format. It also has a built-in safety feature, which I've actually set up, and it was able to accurately understand each of these different components. And you can go ahead and ask it different questions on your codebase. So you can ask it, "What API key do I need and how do I set it up?" So in this case, it tells you that you need the API key from the Grok console, and then you can set it up within an .env file, and it gives you the location. So this is something that has really great understanding of your overall codebase, and it can make really great changes for it.
Now, within the chat interface, you have the ability to access actions. These are different types of actions you have: from finding different documents, explaining files, you have the ability to fix where you can debug a certain section of your code; you have the ability to write different tests; and you also have it so that you can upload and select different documents that you want to work with. Then you have the "Add context," where you can add individual files from your workspace, or you can add folders from the entire codebase itself within the workspace. And another cool thing is that this gives you the ability to conveniently add recently opened files. In this case, we opened the .env file, so you can easily select it, or you can go ahead and add documentation from recently available docs that you have.
To showcase the contextual relevancy feature of Augment, I simply asked it, "Could you please find any errors across my workspace, and could you please resolve them for me right away?" You can see that it found an API key security issue from the API or for the environment variable file. We're going to go ahead and apply this so that it fixes this. There's a missing error handle in the create config file, so we can go over to this file, and we can actually review through it. And once we approve of it, we can click on the "Apply" button, and you can see that it made a diff edit, and you can accept all of them actually by clicking on the "Next" button over here to see all the changes that were made, and then you can click on "Accept." Now, for each edit, you can individually click on the file, and then you can actually visualize where the changes are. So in this case, you can see the addition of the new code, a part of this block, and you can either accept it or reject it. And for each change it made to that individual file, you can see all the different changes made, and then you can individually accept or reject it, or you can go ahead and just simply go ahead and accept all changes.
But guys, this isn't even the best feature. I get it—it is really amazing that Augment can give you full contextual relevancy across your large codebase, and it's something that many other coding agents can't do. But this next feature is just truly insane. This is the configuration file for my web crawler, where I gave it the configuration to crawl with the base URL, the different sorts of dimensions I wanted to crawl on, as well as the extension. But what I'm going to do is showcase the feature of Augment, which is the next-edit feature. This helps you flow through complex changes across your whole codebase effortlessly. So in this configuration file, if I am to make an edit over here, and if I am to add something, it's going to be able to upgrade the whole library; it's going to make the schema changes that are necessary, as well as it's going to fix up all the repetitive tasks like refactoring across the overall codebase, which is just insane.
To showcase this live in action, we're in the configuration file, and what I'm going to do is go all the way to the top and change this action. I'm going to simply go ahead and rename this. This is just a small little edit that would be quite tedious to go over to your overall codebase and change all the configurations. I'm simply renaming it to "promptInput," so it's a bit more clear. And you will see that the next edit will provide you the changes of where you need to delete things or make the necessary changes. And if you want to go ahead and change this action, you can then go ahead and type in Control or Command semicolon so that you can use the next-edit feature, and you can go ahead and make the necessary changes that next-edit makes. So you can then go ahead and delete the trailing bracket and fix it with the next-edit feature, and you can see that it'll make the necessary changes across all the necessary files, all the necessary different folders across your workspace. And that's the capability of this next-edit feature. So you can see how helpful this is in terms of managing large codebases with AI. Usually, there's a context window problem with most large language models, and with coding agents not configured properly, you can't make the necessary changes. But this is a great way for you to fix things, have it explain things, autocomplete code for you if necessary. And that is the gist of what you can do with Augment Code.
If you like this video and would love to support the channel, you can consider donating to my channel through the Super Thanks option below, or you can consider joining our private Discord where you can access multiple subscriptions to different AI tools for free on a monthly basis, plus daily AI news and exclusive content, plus a lot more. In essence, Augment Code automates large-scale code updates by detecting patterns, applying context-aware changes, and suggesting fixes across your codebase instead of manually searching and replacing things. And thanks to the help of AI, this is going to expedite your software engineering tasks quite rapidly. So I definitely recommend that you try this out. With all the links that I used in today's video in the description below, this is truly a game changer, as this is something that you can access for free. So definitely recommend that you try it out, with all the links that I use in today's video in the description below. Make sure you subscribe to the newsletter, follow me on Patreon so that you can join our private Discord, follow me on Twitter to stay up to date with the latest AI news, and lastly, make sure you guys subscribe, turn the notification bell on, like this video, and please take a look at our previous videos, cuz there's a lot of videos on this channel that will definitely benefit you in different areas in the AI space. But with that thought, guys, have an amazing day, stay positive, and I'll see you guys fairly shortly. Peace out, fellas.