Speaker 1: Remember the framework called Cloud Engineer? Well, the developer behind the project, Petro, is back again and has introduced a new framework revolving around the new DeepSeek R1 model, which is supporting its reasoning capabilities to get powerful generations. It's called DeepSeek Engineer, and it is a new powerful coding assistant app that integrates the new reasoning model to process user conversations as well as generate whatever you request it with high quality. Through an intuitive command-line interface, it can read local file contents, it can create new files, apply diff edits to existing files in real-time, and so much more. Now, I already made multiple videos exploring the capabilities of this new DeepSeek R1 model, and I definitely recommend that you watch them with the links in the description below. But to summarize what this model is about, it's an open-source revolutionary large-language model that outperforms proprietary giants like OpenAI's O1, GPT-4 Omni, as well as Anthropic's Cloud 3.5 Sonnet in various benchmarks like coding, mathematics, you have multilingual tasks, and so many others. It's licensed under the MIT license, and it's affordable, accessible, and it supports 128k context length. Now, I definitely recommend that you subscribe to my newsletter because I am going to be constantly posting different sorts of content here that will summarize my videos, so definitely give this a subscribe if you haven't already. But essentially, you can see that this model is definitely quite revolutionary, and it's the first open-source model to be outperforming a state-of-the-art model in various categories. And this is why today, we're going to be showcasing another agent coder that is going to be using this model to help you out in various ways related to coding. Ever wanted to automate your AI workflows and boost your productivity? Well, now you can with Zapier. Zapier connects over 7,000 apps, including your favorite AI tools, to automate the work you've already been doing. Whether you're managing code deployments, or automating AI-powered insights, or setting up workflows to scale your business, Zapier is here to help you focus on the creative, high-level tasks. Did you know, Zapier saves users an average of 10 hours a week? That's 10 hours you can put back into your AI projects, you can put it back into research, startups, or whatever else you love doing. I've set up Zaps personally that automate complex AI workflows, saving me countless hours. You can connect tools like AI training models to data processing apps, or even schedule automatic reports for your AI project. For example, I set up a workflow that automatically saves email attachments to drive with no manual downloading of attachments and organizing them. Zapier is perfect for simplifying your workflow. Once you set it up, it works automatically. Want to automate AI workflows and get time back? Sign up for Zapier today and get a free 2-week trial. Click the link in the description below to get started. But with that thought, let's get right back into the video. Here's a preview of DeepSeek Engineer. You can see that it takes the prompt and it goes ahead and plans. And as you all know, DeepSeek R1 is a reasoning model, which takes the steps to constantly use a chain of thought to work through and generate something with a plan that's quite detailed. And you can see that it is working on crafting up a detailed plan to execute the task, which is to create this game. So once it has generated, you can see then you can open this up and now you're going to be able to see that the game is fully functional, similar to like a portal game and a snake game combined where you got to go ahead and get the objective of collecting these different items. And you can use these portals to basically play through the game. But essentially, this is just a really basic example of how it works as a nice looking interface that can take your request, use the reasoning steps to generate your response. And you also have the ability to have it generate multiple files and make changes to your codebase to execute whatever you requested to do. Now, before we get into installing DeepSeek Engineer, there's a couple of things that you would want to keep in mind, which are the features. You have the ability for an API client configuration, which configures the DeepSeek API client for like a GPT-like completion. You have data models that uses Pydantic for TypeSeek handling. You have system prompts, a helper function that includes to locate your local files, create files and whatnot. You have the ability to add different paths as well as quickly find folder contents. You have a conversation flow as well as an interactive session that you can use within your terminal. Now, to install this locally, there's a couple of prerequisites that we'll need beforehand. So make sure you have Git installed. This is going to be used to help us clone the repository locally on a computer. Make sure you have VS Code to use this interface for the terminal as well as configuring the .env file. And lastly, you will need a DeepSeek API key that's linked to a billing account that has been topped up already. So once you have these prerequisites fulfilled, go over to the GitHub repository, which I'll leave a link to in the description. Click on this green button. Click on copy the link to clipboard. Scroll back down to the get started tab. And what I want you guys to do is open up your command prompt. Now, once you have this opened up, type in git clone and then paste in the link and click enter. This will clone the repository locally. And before you actually even install anything, what you're going to need to do is open up VS Code. I want you to open up the DeepSeek engineer directory. And once you have opened it up within VS Code, right click on the left hand panel. Click on create a new file and then name it .env. You're going to basically go ahead and set your environment variables. You want to go here. Copy this section, which is the DeepSeek underscore API underscore key equals your API key. So go ahead and copy this first section. Paste this in. And then you want to go ahead and paste in your API key that you can retrieve from the DeepSeek platform. And once you have pasted in your API key, you can go ahead and click file and then click save. And then you can go ahead and minimize this for now. Now, what I want you guys to do next is install this using pip. You can also use UV to set up the virtual environment. We're going to go ahead and copy this pip install command. Go back into our command prompt. Type in cd deepseek-engineer to get into the directory of DeepSeek engineer. And then we're going to go ahead and paste in the pip install command. This is going to start fulfilling and installing all the requirements that are needed. And then afterwards, we can go ahead and use the python3 main.py command to start this up. And there we go. We now have this loaded up and we can start interacting with the DeepSeek engineer within our command prompt. Or you can either do this within VS Code by going over to the terminal. And then you can create a new terminal. And then you can start this up with the same python3 command. Or if you're on Windows, you would want to use the python main.py command. So let's start off by generating something really, really basic. Let's say, hey, help me create a basic front end for a SaaS website. We're going to go ahead and send this in. And you'll see that the assistant is going to start working on creating a plan. It's going to then create the path, which you see over here. It's creating the index.html file. It's going to use and create CSS for the styling and then JavaScript for the functionality of this app. But you can see right now, it is working on writing up all the code that is necessary. And it is now creating and updating the file over here. And I'll be able to access it once it has finished. And there we go. We have this SaaS website that was developed. It was able to build out the front end. Obviously, this is something that it can do quite easily. But we're going to start on working on something that's a bit more intricate. So you get a better idea of what you can do with this. So now what I'm going to have it do is create a local finance investment manager app. And this is by using Python. And I wanted to basically have it so that it tracks different investments. And then you can have it so that you can edit, remove different entries of whatever stock you bought, as well as having it so that you can place in your bonds as well as other investments. And there we go. We have our basic Python app. And this is something that's saved within the directory. So I've now went along and open this up. And there we go. We have our simple Python app, which is where you can track your investments locally. So you can see over here, you have your asset type. Let's say we bought Apple. Let's say we bought around $2,000 worth of this at the price of 180. I don't know what's it at right now. But let's go ahead and add this. And you can see that you're going to be able to track your stocks within this app. Now, next up, I'm going to have it create a simple web app. And essentially, this is going to be a tasker app that tracks different types of tasks. And it then portrays whatever tasks that you upload into a calendar so that you can see what sort of things are coming up for your week. And essentially, this could be like a really nice app that you can use to help you track out all your busy events that could be happening throughout the month or throughout the year. And there we go. This is the task calendar app that it was capable of generating. You have your tasks that you can upload on the left-hand panel, such as creating a YouTube video. And then you can go ahead and add this task. And you're going to be able to visualize it within the calendar. You can see the total amount of tasks. And then you can also go ahead and checkmark this so that you have completed this task. And in this case, you can also visualize your pending task, completed task, and all your task over here. And this is a great way for you to easily track different tasks. And this was all created with the help of DeepSeek Engineer. 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. Now, that's basically it for today's video on DeepSeek Engineer. I definitely recommend that you try taking a look at this with the link in the description below. It's not like your regular front-end app builder. This is more of an intricate way for you to work with DeepSeek to help you reason and plan out different developments with multiple different files for web apps, as well as for local projects that you can get started with easily from scratch. So definitely recommend that you take a look at this with all the links that I used in today's video in the description. Make sure you follow me on the Patreon. Make sure to subscribe to the newsletter. Follow me on Twitter for daily AI updates. And lastly, make sure you get subscribed, turn on the notification bell, like this video, and please take a look at our previous videos, because there's a lot of content that you will definitely benefit from. But with that thought, guys, have an amazing day, spread positivity, and I'll see you guys fairly shortly. Peace out, fellas.
Generate a brief summary highlighting the main points of the transcript.
GenerateGenerate a concise and relevant title for the transcript based on the main themes and content discussed.
GenerateIdentify and highlight the key words or phrases most relevant to the content of the transcript.
GenerateAnalyze the emotional tone of the transcript to determine whether the sentiment is positive, negative, or neutral.
GenerateCreate interactive quizzes based on the content of the transcript to test comprehension or engage users.
GenerateWe’re Ready to Help
Call or Book a Meeting Now