hi there and welcome to mind Studio after creating an account and naming your workspace you'll see your worksspace home screen where you can begin using AI right away using your workspace assistant your workspace assistant is the default AI app that opens when you enter your workspace to use the assistant simply type something into the chat you can customize your assistant by clicking on the gear icon then select edit AI this will take you into the AI editor where you can modify your app in many different ways at the top of your workspace you'll find several other controls on the top right you can access your workspace settings by clicking on your workspace name then clicking on the gear icon from here you can update your workspace name company info and membership plan as well as view some analytics set usage limits and invite new members to your workspace to adjust your profile settings click on the profile picture at the top right here you'll be able to update your username email password and more lastly you can access additional Resources by clicking on the question mark icon next to your profile picture from here you can access our learn page our documentation or talk to our support team via a chat the plus button at the top left will open up a new tab in your workspace from here you'll be able to create use and manage all of your AI applications if this is your first time using mine Studio the space will be empty to create a new AI click on the create AI button which will take you to the new AI setup screen from here you can explore all kinds of templates for various use cases or you can start from scratch we'll be building a blog post generator from scratch in this video when starting from scratch it's recommended to use the generate prompt option this option will use M Studio's AI prompt engine to assist us with the initial setup of our AI application after tapping on the next button it will ask us about the AI that we're trying to build simply describe the AI in a couple of sentences using natural language when writing instructions for the AI it's good practice to refer to the AI as assistant and to refer to the end user as human after clicking generate you'll be taken to the AI editor for the first time on the left hand side you'll find the Explorer which is where all of the resources used in building your AI applications can be found this includes resources like custom data sources which are files that you can upload for the AI to query and reference functions which allow you to run JavaScript code in your AI workflows user inputs which are various kinds of forms that an end user can fill out to provide contextual information to the AI and workflows which is where you'll create your automations and spend most of your time building in the center of the Ed is the main workspace this area will change depending on the resource you're working in you can see here that we're currently in the prompt tab of the mainflow workflow The Prompt tab is where you'll write your system prompt this is a simple set of instructions that lays the foundation for your AI workflow you'll notice the system prompt for our workflow was already written for us by the AI prompt generator review it quickly to make sure the information is correct or write additional instructions before moving on to the next step as we build out our AI application we'll need to make sure that we use the right AI model for our specific workflow we can configure this from the model settings tab from here we'll be able to select the underlying model for our AI application you can see that we use CLA 3 hu by default if we click the selector we'll be able to choose from a wide variety of of AI models each model has its own strengths and weaknesses and it's important to consider things like cost speed context limits and quality When selecting an AI model for our app for this app I'm going to go ahead and use GPT 3.5 we can also adjust things like the temperature which is the amount of Randomness in a model's response and the max response size which is the maximum length of an AI response typically it's recommended that you start in the center of these two sliders and adjust the values as needed the automations canvas is where you will be building out the automated actions of your AI application automation workflows are set up into three key Parts the start block which initializes the AI workflow the Terminator block which determines the final user experience and marks the end of the workflow and everything in between in the canvas you can scroll up and down left and right and if you ever get lost you can use the home icon to Center all of the blocks in your canvas you can add new blocks to your automation workflows by clicking on the plus button in the canvas this will open up a drop- down menu where you can add all kinds of blocks to your AI workflow the drop down is separated into different sections that correspond to various types of automations since we're building a simple content generator we'll want to add a couple of blocks to our AI workflow the first is a way to let our end user specify a topic for the AI to write a blog post about we can do this by adding a user input block you'll notice that after adding the block there are no user inputs in the configuration area on the right we can create a new user input by clicking on the plus button next to the user inputs folder there are all kinds of inputs that collect different kinds of information from our users like multiple choice file uploads or text inputs you can see a preview of the user input on the right for this workflow we're going to use the long text input all inputs are saved as variables that we can recall anywhere in our workflows to configure this user input we'll save this input as a variable named topic and we'll change the label text now we can return to our workflow and add the input to our block click on the user input block then click on the plus button and select the input that we just created blocks can also be dragged around the canvas to create space next we'll need to use AI to generate some written content for our blog post based on the topic provided to do this we'll add a generate text block this block sends a message to a large language model and returns a response to configure the generate text block we can simply enter the prompt we'd like to send to the AI you'll notice that in our generate text prompt we are recalling the variable topic by using these double curly braces additionally we might also want to utilize AI to come up with ideas about the topic provided this will help it generate more effective text content for our blog post to do this we can add another generate text block and then inside of this block we can type in our prompt to generate ideas in this case we won't want to display the text to the user instead we're going to go into the settings area and we're going to want to save this response as a variable we'll call this variable ideas additionally we can use multiple models within our AI workflow to make sure we're using the right model for the right automated task to do this we can go down into model settings and select the model we'd like to use for this specific block from here we'll also need to adjust the temperature and the max response size back in the final generate text block that writes the blog post we can also include this ideas variable so that it makes sure to incorporate the ideas now our basic workflow is complete let's quickly summarize we start the workflow and initialize it we collect a user input that asks our end user to provide a topic we then have the AI come up with some ideas surrounding the topic provided and finally we generate the blog post for the Terminator blog block we will end in a chat so that the user can continue chatting with the AI after the blog post is written before we publish our app we should make sure that there are no errors in our project we can check this by looking at the errors tab let's also take a look at a preview of our AI application we can do this by tapping on preview and then clicking on preview draft this will open up a tab where we'll be able to see what the AI will look like to our end users on the right hand side we can also open up a debugger which will showcase all of the actions taken within our AI workflow let's go ahead and test out our AI as it processes the blog post you'll notice that all of the steps are appearing on the right hand side as they're generated you can see each action step by step the thread has been started a user input was collected a background message was used for AI processing and then the blog post was written you can see the final output of the blog post here and we can continue to scroll down after tapping next the user will also be able to continue chatting with the AI back inside of our editor tab we can begin finalizing our project if we go to the Explorer Tab and click on the name of our AI in the top folder we'll be able to edit some settings around our AI from here we can give our AI a name and a short description we'll also be able to go down into several other sections that allow us to create onboarding flows add some icons in Media style our AI applications and set usage limits each of these settings are discussed in depth in our documentation and other videos now that we've set up our AI settings we can go ahead and publish our AI application to do this we simply tap on the publish button at the top right and then we can click on a link to open up our published application in our AI browser now we'll be able to use this AI within our workspace whenever we'd like when the tabs are closed and we're in our home screen we can access our AI by again tapping on the plus button and then we'll find our AI in the list of published AIS to open the AI we can simply click on the AI and it will open within the tab we can then edit our AI by clicking on the gear icon and then selecting edit AI this will take us back into the editor mind Studio allows you to create many kinds of AI app applications there are many features that were not covered in this getting started video but we hope that this gives you a foundation for beginning to build your own AI applications for additional resources make sure to check out our YouTube channel and our documentation to learn about other features on mind Studio thanks so much for watching and we'll see you next time