Make Applications using Free AI Software Engineer on Your Computer: Devika

Key Takeaways

this is amazing now we have the first open-source AI software engineer when we give a prompt like this Implement Conway's Game of Life in Python using py game it is able to list all the steps browse the internet for more information and able to create files it is going through step by step and able to run the game that's exactly what we're going to see today let's get [Music] started hi everyone I'm really excited to show you about DEA agent dig AI software engineer this is completely open source so you can try this on your computer even using open source LOD language model using olama you can see it supports Cloe 3 gp4 GPD 33.5 and local models via olama Advanced AI planning and reasoning capability contextual keyword extraction seamless web browsing code writing in multiple languages Dynamic agent State tracking and visualization natural language interaction via chat interface which means on the left hand side you will chat with the agent and right hand side you can see the progress I'm going to take you through step by step on how to install this on your computer and test this but before that I regularly create videos in regards to Artificial Intelligence on my YouTube channel so do subscribe and click the Bell icon to stay tuned make sure you click the like button so this video can be helpful for many others like you in this we are going to see how to configure this then how to set up the frent then how to set up the back end in this we are going to run two softwares one is flask written in Python that is the back end so we will keep that running and in the front end we will use BN it's a JavaScript based front end application so that's what we see in our browser so first step git clone and the GitHub URL SL Dava and then click enter next navigate to the Dava folder now cond create hyph and Dava Python 3.11 and then click enter Now cond activate DEA and then click enter next pip install hyph requirements.txt and then click enter ENT this will install all the required dependencies or the packages now if you open this code folder in VSS code you can see various files among that we need to change config TL just open it it will look like this so here you got API keys so you need to enter your Bing API key if you're using Cloe you can use clo API key if you're using open AI you can use open AI API key so replace this accordingly one difference is that for Bing search API key you might need to follow different steps so you got the Bing search API page there you can click try now there you will be present with a page like this enter the subscription Resource Group or you can create a new one then provide a name for the instant select the pricing tier so you can choose the first one for basic tests then confirm and you can create it once after you create you will have a page like this where you will have keys and endpoint click on that and there you should be able to copy your API key so this Bing API is used for browsing the internet or searching similarly if you go to platform. openai you should be able to create API keys for open aai in console. anthropic you can get API keys for clo once after you modify this save the file you don't need to use Cloe or open AI at the same time you can use one of those whatever you prefer now in your terminal python dear. py and then click enter this will automatically start the backend flask application so keep the terminal running let's open a new terminal navigate to UI folder CD UI now we need to install bun to set up your friend end application it is using JavaScript so I'm going to copy this code to install bun and paste it here and then click enter now it's installing butt and it has been installed now inside the UI folder as you can see here just type BN install and then click enter I've already installed before so it says already installed now Bun Run Dev and then click enter this will automatically start the front end now you can open this URL and here is the interface you can go up and create a new project let's say test project now we need to choose the model we want to use I'm going to use gbd4 so now going to ask Implement convey Game of Life in Python using py game and then click Send now you can see on the left hand side you can talk to the AI engineer and here you can see it's performing the request it's listing all the steps on the left hand side here is my step-by-step plan set up a python development environment install Pi game research about cony's Game of Life so it's browsing in the right hand top corner you can see that now it's going through the game development tutorial now it's researching about improving performance in P game I can see at the bottom it's trying to run in the back end you can see all the request coming through now in the readme file I can see the list of things to follow how to run ensure you have the python installed on your system using pip install hyph requirements.txt then python run so you can find all these files in the projects folder so this is your folder structure go inside projects so CD projects and inside that I created a project called test that's what we defined in our user interface then click enter now inside this file you can see three different files that's what we saw on the screen Game of Life read me and requirements.txt now I'm going to run pip install hph requirements.txt this will install the P game package and all the dependencies so I can see an error here so it's because of the requirements.txt file so if I open the requirements.txt file I can see the version here so I'm going to delete it because it's using older version so that's the reason it's getting failed so going to delete that and do the same step again pip install hyen requirements.txt and then click enter this is this is all depending on the large language model you're using not because of this tool now going to run the python gameof life. py file and you can see the game got started but I don't see anything but this is depending on the L language model which we are using the more powerful the lar language model is then definitely this code is going to work so overall this AA engineer is able to perform the task be requested you can also follow the documentation to set up your open source Lun language model using o I'm really excited about this I'm going to create more videos similar to this so stay tuned I hope you like this video do like share and subscribe and thanks for watching