1 of 100

Documentation

Overview

Welcome to MindStudio

In just a few minutes, you’ll learn how to launch, customize, and deploy your own intelligent tools—no coding needed.

Explore MORE...

Building AI Agents

Editor Overview

Get a high-level overview of the MindStudio AI Editor

After creating a new AI Agent, you’ll land in the MindStudio Editor. The editor is made of two key areas: the Explorer and the Navigator (commonly referred to as the Workspace Area).

Explorer Tab

On the left you’ll find the Explorer Tab. This is where you'll find all of the resources used to build your AI Agents.

Creating Workflows

AI Agents are just workflows.

What is a Workflow?

Workflows are sequences of automated actions that your AI Agents follow when they are run.

Designing Workflow Automations

After opening the Automations Tab, you’ll see the Automations Canvas which displays all of the individual actions that your workflow will execute when run. These are represented by Blocks on the canvas.

Best Practices for Designing Workflows

Plan Before You Build: Outline the steps your workflow will perform.
Use Helpful Variable Names: Name variables descriptively to make the workflow easy to debug.
Test as You Go: Regularly test each block to identify and resolve issues early. To test out the specific output of blocks that use AI, you may use the Profiler to compare results.

Automations Canvas Controls

The canvas is infinitely scrollable in all directions. At the bottom of the canvas, you’ll find several controls to help you navigate and annotate the canvas as you design your Workflow.

Panning Around the Canvas

Using a mouse:

Scroll to pan up and down the canvas.
Shift + Scroll to pan left and right across the canvas.

Using the Pan Tool:

Click the Pan Tool icon or use the H hotkey to activate the Pan Tool.
Click and drag in any direction to pan around the canvas. NOTE: You will not be able to select blocks while the Pan tool is activated.
Click the Select Tool icon or use the V hotkey to deactivate the Pan Tool.

Zooming In & Out

Using a mouse:

On Mac, use CMD + Scroll to zoom in and out.
On PC, use CTRL + Scroll to zoom in and out.

Using the Zoom Controls:

Click Zoom In icon or use the + hotkey to zoom in.
Click Zoom out icon or use the - hotkey to zoom out

Centering the Canvas

If you navigate away from your blocks and can no longer find them, you can reset and center all of the blocks on the canvas by clicking the Reset view icon or using the R hotkey

Auto Arrange Blocks

If your workflow becomes long and complex, you may want to consider tidying it up using Auto arrange. This tool will align all blocks vertically on the canvas.

Adding Notes to the Canvas

You can annotate the canvas using the Note Tool to add text notes or label groups of blocks on.

Using the Note Tool Controls:

Click on the Note tool icon to create a new note
Select the color of your note
Click on the note and add text content to your note
Add an optional label to the top of your note. If no text is added in the label, then it will not show a label.

Add a New Block to the Workflow

Between Existing Blocks:

Select the + button between to connected blocks to open up the Block Menu.
Select the block that you’d like to add.
The new block will be automatically connected between the two blocks.

Anywhere on the Automations Canvas:

Right-click anywhere on the canvas to open the Block Menu.
CTRL + Click anywhere on the canvas to open the Block Menu.

Deleting a Block

Click on the delete icon at the top-right corner of any block.
Confirm the delete action.

Key Components of a Workflow

The Start Block

This block initializes the workflow. It can be triggered on demand or on a defined schedule, and allows you to define launch variables, which provide dynamic values that are passed through the workflow.

Workflow Blocks

There are many kinds of blocks that you can add to your workflow. All blocks will have different configuration options depending on the block that you select.

Types of blocks include:

AI Tools: Generate text, generate images, analyze image, etc.
Context Blocks: Gather context for the AI. Context is saved to a variable.
Routing Blocks: Add conditional branches to route the workflow in various ways. </aside>

Terminator Blocks

This block marks the end of the workflow. It has customizable end behavior, such as sending email notifications or returning a structured output.

Defining and Calling Variables

Variables in MindStudio are dynamic placeholders that store data during workflow execution. They allow you to pass information between blocks and workflows seamlessly.

Defining Variables

Launch Variables: These are defined in the Start Block of your workflow.
Runtime Variables: Some blocks, such as Generate Text Blocks or User Input Blocks, can generate new variables while the workflow is running. For Example, after performing a Google Search, the block can store the results in a variable called google_result.

Calling Variables

To use a variable in any block, reference it using double curly braces: {{variable_name}}.

Generate Text prompt example calling a variable:

Analyze Image Block

Analyze an image URL based on text instructions.

The Analyze Image Block processes and analyzes an image URL based on provided text instructions. It generates a text response based on the analysis that is assigned to a variable.

Configurations

Prompt

Provide text instructions for analyzing the image. Use {{variables}} to make the prompt dynamic based on a previous step in the workflow.

Example Prompt:

Identify the objects in the image and describe their arrangement. Focus on colors and sizes.

Image URL

Specify the URL of the image to be analyzed. Use a {{variable}} to dynamically reference image URL generated or fetched earlier in the workflow.

Output Variable

Creates a new variable and saves the generated test response to it. Enter a variable_name to store the response for later use in the workflow.

Model Settings

Choose the AI Model that you’d like to use for the image analysis. You may adjust the model’s Temperature and Max Response Size.

Writing Image Analysis Prompts

Creating a well-crafted prompt is essential for accurate and meaningful image analysis. Follow these guidelines to ensure your prompts guide the AI model effectively:

Be Clear and Specific: Clearly describe what you want the AI to analyze or identify in the image. Focus on specific elements or features relevant to your use case.
Example: "Identify the main objects in the image, including their colors, sizes, and positions.”
Define the Objective: Explain the purpose of the analysis to provide context for the AI. This helps tailor the response to your needs.
Example: "Analyze the image and describe how the objects are arranged for interior design recommendations.”

Chat Block

Start a chat session with AI

Configuration

Chat Settings

Checkpoint Block

Allow for user review of a specific step in the workflow.

Overview

The Checkpoint Block displays a prompt to the user, allows them to revise variables, accept or reject certain steps in the workflow and route them accordingly.

Configurations

Create Google Sheet

Create a new Google Sheet document in a MindStudio workflow

Configuration

Account

Create LinkedIn Post

Make a new post on LinkedIn in a MindStudio workflow

Configuration

Connect a LinkedIn Account

Create Notion Page

Create a page in Notion

Configuration

Account

The Fetch Slack History block retrieves all messages from a specified Slack channel within a chosen date range.

Configuration

Account

Configuration

Account

Dynamic (Ask at runtime)

This setting will ask a user to login to their Coda account when they run the app.

Get Coda Table Rows

Fetch row data from a table in Coda.

Configuration

Account

Dynamic (Ask at runtime)

This setting will ask a user to login to their Coda account when they run the app.

Jump Block

Route the current workflow to a completely different workflow

The Jump Block enables you to route the current workflow to a completely different workflow. This block is ideal for modularizing workflows or reusing common processes across multiple workflows.

Configurations

List Upcoming Calendar Events

List upcoming events from your Google Calendar

Configuration

Account

Send Email

Send an email to a registered address

Configuration

Account

Send SMS

Send an SMS to a valid phone number

Configuration

Account

Set Run Title

Update the title in the user's history

The Set Run Title block updates the title that appears in the user's run history for a given Run. This makes it easier for users to identify and organize previous interactions, especially when agents are used repeatedly with different content or contexts.

Configurations

Creating Workflows

AI Agents are just workflows.

What is a Workflow?

Workflows are sequences of automated actions that your AI Agents follow when they are run.

Designing Workflow Automations

Best Practices for Designing Workflows

Plan Before You Build: Outline the steps your workflow will perform.
Use Helpful Variable Names: Name variables descriptively to make the workflow easy to debug.
Test as You Go: Regularly test each block to identify and resolve issues early. To test out the specific output of blocks that use AI, you may use the Profiler to compare results.

Automations Canvas Controls

The canvas is infinitely scrollable in all directions. At the bottom of the canvas, you’ll find several controls to help you navigate and annotate the canvas as you design your Workflow.

Panning Around the Canvas

Using a mouse:

Scroll to pan up and down the canvas.
Shift + Scroll to pan left and right across the canvas.

Using the Pan Tool:

Click the Pan Tool icon or use the H hotkey to activate the Pan Tool.
Click and drag in any direction to pan around the canvas. NOTE: You will not be able to select blocks while the Pan tool is activated.
Click the Select Tool icon or use the V hotkey to deactivate the Pan Tool.

Zooming In & Out

Using a mouse:

On Mac, use CMD + Scroll to zoom in and out.
On PC, use CTRL + Scroll to zoom in and out.

Using the Zoom Controls:

Click Zoom In icon or use the + hotkey to zoom in.
Click Zoom out icon or use the - hotkey to zoom out

Centering the Canvas

If you navigate away from your blocks and can no longer find them, you can reset and center all of the blocks on the canvas by clicking the Reset view icon or using the R hotkey

Auto Arrange Blocks

If your workflow becomes long and complex, you may want to consider tidying it up using Auto arrange. This tool will align all blocks vertically on the canvas.

Adding Notes to the Canvas

You can annotate the canvas using the Note Tool to add text notes or label groups of blocks on.

Using the Note Tool Controls:

Click on the Note tool icon to create a new note
Select the color of your note
Click on the note and add text content to your note
Add an optional label to the top of your note. If no text is added in the label, then it will not show a label.

Add a New Block to the Workflow

Between Existing Blocks:

Select the + button between to connected blocks to open up the Block Menu.
Select the block that you’d like to add.
The new block will be automatically connected between the two blocks.

Anywhere on the Automations Canvas:

Right-click anywhere on the canvas to open the Block Menu.
CTRL + Click anywhere on the canvas to open the Block Menu.

Deleting a Block

Click on the delete icon at the top-right corner of any block.
Confirm the delete action.

Key Components of a Workflow

The Start Block

Workflow Blocks

There are many kinds of blocks that you can add to your workflow. All blocks will have different configuration options depending on the block that you select.

Types of blocks include:

AI Tools: Generate text, generate images, analyze image, etc.
Context Blocks: Gather context for the AI. Context is saved to a variable.
Routing Blocks: Add conditional branches to route the workflow in various ways. </aside>

Terminator Blocks

This block marks the end of the workflow. It has customizable end behavior, such as sending email notifications or returning a structured output.

Defining and Calling Variables

Variables in MindStudio are dynamic placeholders that store data during workflow execution. They allow you to pass information between blocks and workflows seamlessly.

Defining Variables

Launch Variables: These are defined in the Start Block of your workflow.
Runtime Variables: Some blocks, such as Generate Text Blocks or User Input Blocks, can generate new variables while the workflow is running. For Example, after performing a Google Search, the block can store the results in a variable called google_result.

Calling Variables

To use a variable in any block, reference it using double curly braces: {{variable_name}}.

Generate Text prompt example calling a variable:

Generate Image Block

Use AI models to generate images in your AI Agents.

The Generate Image Block sends a text prompt to an AI model and returns an image URL based on the description provided.

Configurations

Prompt

Define the text description of the image you want the AI model to generate. Use {{variables}} to make the prompt dynamic based on a previous step in the workflow.

Output Variable

Creates a new variable and saves the image URL to it. Enter a variable_name ****to store the response for later use in the workflow.

Model Settings

Choose from MindStudio’s library of image-generation models, optimized for various styles and quality levels.

Each AI image model may have unique configuration options for fine-tuning outputs, such as additional style parameters, rendering quality, negative prompt or advanced filters. make sure you review the specific settings for each model you select.

Writing Image Prompts

Crafting effective prompts for image generation ensures that the AI model produces visuals that align with your expectations. A good prompt provides clarity, context, and enough detail to guide the model’s output without being overly restrictive.

Tips for Writing Effective Prompts

Be Clear and Specific: Use precise language to describe the desired image. Focus on key visual elements, such as objects, environments, colors, and textures.
Add Context and Atmosphere: Describe the mood, setting, or story behind the image to provide creative direction.
Include Style and Aesthetic Details: Specify the art style or medium (e.g., photorealistic, watercolor, comic-style, abstract). Mention influences like "minimalist," "vintage," or "cinematic glow."

Common Pitfalls to Avoid

Being Too Vague: "Create an image of a tree." → Lacks detail and context.
Overloading the Prompt: Avoid including too many unrelated elements, which can confuse the model.
Leaving Out Key Details: Forgetting to mention the desired style, mood, or setting can lead to generic results.

Example Image Prompt:

Generate a hyper-realistic image of a futuristic cityscape at sunset. The city should feature towering skyscrapers with glass facades reflecting the orange and pink hues of the sunset.
Include floating vehicles with soft neon underlights flying between the buildings, creating a sense of motion.
In the foreground, add a bustling street market with humans and humanoid robots interacting, showcasing a blend of futuristic technology and traditional market stalls.
Use a warm color palette dominated by orange, pink, and gold, with subtle accents of teal and blue in the neon lights.
The sky should be filled with soft clouds, and the setting sun should cast long shadows across the scene.
Ensure the image feels dynamic and alive, capturing both the grandeur of the skyline and the vibrancy of the street-level activity. The overall style should balance photorealism with a slight cinematic glow for a futuristic yet inviting atmosphere.

Example Output (DALL-E 3):

Image CDN Transformation Parameters

MindStudio’s Image CDN lets you dynamically resize and optimize images using URL query parameters. Just add these parameters to your image URL to customize how it's delivered.

Base URL

All images are served from:

To transform an image, append query parameters to the URL like this:

Supported Parameters

`w` – Width

What it does: Sets the width of the image in pixels.
Example: ?w=400 (image will be resized to 400px wide)

`h` – Height

What it does: Sets the height of the image in pixels.
Example: ?h=300 (image will be resized to 300px tall)

`fm` – Format

What it does: Changes the file format of the image.
Supported values:
- webp
- avif

`fit` – Fit Mode

What it does: Controls how the image fills the width and height you provide.
Supported values:
- crop – Crops the image to fit exactly

Tips

You can combine parameters, like ?w=400&h=300&fm=webp.
If no parameters are added, the original image is served as-is.
These transformations are handled automatically and served via a fast global CDN.

Start Block

The first part of any workflow in MindStudio

The Start Block is the first part of any workflow in MindStudio. It initializes your AI Agent, defines how and when the workflow begins, and sets up key variables that are used throughout the workflow.

Configurations

Triggers

Define how the workflow is activated. Triggers can be configured to run:

On-Demand: The default trigger. Requires manual execution.
Scheduled: At specific times or intervals, such as daily at 8:00 AM.
Event-Driven: Based on external inputs like API requests, webhooks, email.

Launch Variables

Variables that are initialized at the start of the workflow and passed through the entire process. Launch variables can be referenced throughout the workflow using {{variable_name}}.

Launch variables can be defined in different ways, depending on the how the workflow is being started. They can be:

Set in the body of an API request
Defined in the settings of a block, if the workflow is being started by another workflow
Defined in the schedule rules (see below) for scheduled runs
Passed via query string parameters to the URL of the agent (e.g., https://app.mindstudio.ai/agents/agent_id/run?variableName=variable_value&anotherVariable=different_value

Run Modes

On-Demand

The default mode for new Agents. These workflows only run when manually triggered by a user or system.

Schedule

The Scheduler is a feature within the Start Block that allows you to set up automated, time-based triggers for your workflow. To enable the Scheduler, change the Triggers > Run Mode configuration from “On-Demand” to “Scheduled”. You may create multiple schedules for a workflow.

Creating Schedules

Define the Schedule using natural language to describe how often and when the workflow should run.
- Examples:
  - "Every day at 8:00 AM"
  - "Every Monday and Wednesday at 3:00 PM"

Browser Extension

Triggered directly via while browsing the web. Enabling users to run Agents on the current page or selected content. Data from the page you are browsing are extracted as Launch Variables through the workflow.

Launch Variables:

url - the URL of the page you are browsing
metadata - A dictionary of metadata including page title, description, and other OpenGraph data when available.
pageContent - All main content extracted from the visible portion of the page. ON sites like youtube, pageContent is also used to extract the video transcript.

Email Run

Triggered by sending an email to a unique, agent-specific email address. This mode is useful for processing inbound email data or automating responses. Content from the email are passed as Launch Variables through the workflow.

Launch Variables:

from – The sender’s email address
subject – The subject line of the incoming email
message – The full text content of the email

Webhook Run

Use this mode to start workflows automatically in response to external webhooks. The webhook endpoint is generated for each workflow and can be connected to other tools or services. Data sent in the webhook payload (e.g., JSON body) is mapped to Launch Variables, enabling full dynamic control over workflow behavior.

Run Workflow Block

Execute a sub-workflow within a parent workflow

REQUIRED: Before a sub-workflow can be used in the Run Workflow Block, the sub-workflow must be configured with:

Launch Variables in the Start Block
Structured JSON Outputs in the End Block.

If these are not configured, then the workflow will not run properly.

The Run Workflow Block allows you to execute a separate sub-workflow within your main workflow. This block is ideal for running sub-processes and reusing common processes across multiple workflows.

Configurations

The Run Workflow block allows you to execute a separate sub-workflow from within the current workflow. It’s ideal for reusing logic, modularizing large flows, or performing iterative processing with dynamic inputs.

To function properly, the selected sub-workflow must be configured with:

Launch Variables in the Start Block
Structured JSON Outputs in the Terminator Block

If these are not present, the workflow will fail to execute.

Agent

Select the AI Agent that contains the workflow you want to run. Use the dropdown to choose from your available agents.

Workflow

Choose the specific sub-workflow from the selected agent. Only workflows that include Launch Variables and JSON Outputs will appear here.

Parameters

Launch Variables

Assign values to the sub-workflow’s input variables. These can be static or dynamic using

{{variables}}.

Example:

customer_name → {{userName}}
order_id → 12345

Output Variables

Map the structured outputs returned from the sub-workflow into variables in the parent workflow.

Example:

customer_profile → customer_info

Iteration Settings

Use this section to run the selected workflow multiple times — once for each item in an input list.

Iterator

Choose how items are extracted for iteration:

Auto-Extract (Default): Uses the Extraction Prompt to pull items from unstructured input.
String Array: Treats the input as a delimited string (e.g., comma-separated list).
JSON Array Input (Advanced): Expects a valid JSON array; each item becomes a separate run.

Input Data / Input Array

Specify the input variable that contains the array or string to iterate through. For example: {{user_email_list}}

Extraction Prompt (only for Auto-Extract)

Provide an instruction for the thing you'd like to pull from the input variable. Each extracted item is stored as a variable called {{item}}.

Ex: Extract all emails

Output: The variable {{item}} is the email that is extracted

Item Separator (only for String Array)

Choose how to split the input string:

Comma
Semicolon
Pipe
Custom

Ex: If the input string is Dog, Cat, Fish , then select the Comma option

Output Variable

The result of each sub-workflow run is stored in a JSON array at this variable name.

Example:

batch_results

Execution Mode

Choose how runs are processed:

Parallel: Runs all iterations in batches (faster)
Sequential: Runs one at a time (slower)

Error Behavior

Set how the workflow should respond to errors:

Fail: Stop all runs on first error
Ignore Failed Runs: Skip failed items and continue

Retry Attempts

Define how many times to retry failed runs. Use cautiously to avoid unintended costs.

Preparing the Sub-Workflow

Launch Variables:

Define the launch variables that the parent workflow will provide. Go to the sub-workflow you'd like to run and add Launch Variables in the Start Block of the sub-workflow.

Example Launch Variables:

customer_name: The parent workflow will pass the customer name down to the sub-workflow.
order_id: The parent workflow will pass the ID of the customer's order down to the sub-workflow.

Structured Outputs:

Define the Return Data that the sub-workflow will return to the parent workflow. Add these variables to the sub-workflow’s Terminator Block under Return Data -> JSON Output.

Example Return Data:

customer_profile: Returns customer details retrieved from the sub-workflow.
order_status: Returns the status of the processed order from the sub-workflow.

Note: If a workflow lacks these configurations, you will see an error message:

"The selected workflow has not been configured with launch variables and structured outputs."

Best Practices

Ensure Variable Alignment: Make sure that required variables in the sub-workflow are properly assigned with values during execution.
Optimize Your Workflows: Use this block to simplify complex workflows by breaking them into manageable, reusable components.
Test Input and Output Mapping: Verify that all variables are passed correctly between the parent and sub-workflows.
Maximum Iterations Limit

{
  "fullName": "Dmitry Shapiro",
  "linkedin_internal_id": "3203",
  "first_name": "Dmitry",
  "last_name": "Shapiro",
  "public_identifier": "dmitry-shapiro-a2b1",
  "background_cover_image_url": "https://media.licdn.com/dms/image/v2/D5616AQEyFJsY9z6FPg/profile-displaybackgroundimage-shrink_200_800/profile-displaybackgroundimage-shrink_200_800/0/1707167096861?e=2147483647&v=beta&t=ZrJuQxSfNaPug6uf6_2CGumdQfl_v77pZmIy6LxOGIY",
  "profile_photo": "https://media.licdn.com/dms/image/v2/D5603AQHhQ4-UhGgpqA/profile-displayphoto-shrink_200_200/profile-displayphoto-shrink_200_200/0/1708916323677?e=2147483647&v=beta&t=pgJKywpBd1uCcVId7ZXTA6OPHGO3VEyQxS36iXc3iMw",
  "headline": "",
  "location": "San Diego County, California, United States",
  "followers": "30K followers",
  "connections": "500+ connections",
  "about": "",
  "experience": [
    {}
  ],
  "education": [],
  "articles": [
    {
      "link": "https://www.linkedin.com/pulse/world-models-jepa-next-evolution-ai-architecture-dmitry-shapiro-1xcsc",
      "image": "https://media.licdn.com/dms/image/v2/D5612AQHsxV3R2BAWYw/article-cover_image-shrink_720_1280/article-cover_image-shrink_720_1280/0/1739115917775?e=2147483647&v=beta&t=ZHmxIvAX38zXu9p3clVJkrh6btoUJX94WxJ-iPeOtNY",
      "title": "World Models and JEPA: The Next Evolution in AI Architecture",
      "author": "",
      "published_date": "Feb 9, 2025"
    },
    {
      "link": "https://www.linkedin.com/pulse/qwen-25-alibabas-breakthrough-open-source-ai-dmitry-shapiro-8tepc",
      "image": "https://media.licdn.com/dms/image/v2/D5612AQGUy2CzCYkg0w/article-cover_image-shrink_720_1280/article-cover_image-shrink_720_1280/0/1738161461626?e=2147483647&v=beta&t=D_JXXQhMwfRbN8yQfEMTR8wYDCPUcSO0SDGVY4kdLqk",
      "title": "Qwen-2.5: Alibaba's Breakthrough in Open-Source AI",
      "author": "",
      "published_date": "Jan 29, 2025"
    },
    {
      "link": "https://www.linkedin.com/pulse/deepseeks-openai-connection-unraveling-training-dmitry-shapiro-tdj3c",
      "image": "https://media.licdn.com/dms/image/v2/D5612AQFUOE0mb_JQAw/article-cover_image-shrink_600_2000/article-cover_image-shrink_600_2000/0/1738123130175?e=2147483647&v=beta&t=1rI8TcN3dklkag2S8vJuzCzTEpz4iYhIaQfN-1EEVpg",
      "title": "DeepSeek's OpenAI Connection: Unraveling the Training Relationship",
      "author": "",
      "published_date": "Jan 29, 2025"
    },
    {
      "link": "https://www.linkedin.com/pulse/transforming-marketing-function-ai-automation-dmitry-shapiro-o6hcc",
      "image": "https://media.licdn.com/dms/image/v2/D5612AQHzGnV9th3CRQ/article-cover_image-shrink_720_1280/article-cover_image-shrink_720_1280/0/1720672865633?e=2147483647&v=beta&t=UUqASUKVj_Q9Rar_F58GvMxZZEwh5nsgLFVly3FTVUY",
      "title": "Transforming the Marketing Function with AI Automation",
      "author": "",
      "published_date": "Jul 11, 2024"
    },
    {
      "link": "https://www.linkedin.com/pulse/most-sales-people-suck-can-we-get-them-perform-better-shapiro-5p4cc",
      "image": "https://media.licdn.com/dms/image/v2/D5612AQFovihmo_uxVw/article-cover_image-shrink_720_1280/article-cover_image-shrink_720_1280/0/1720287771219?e=2147483647&v=beta&t=VkNixDDA6pwlJtM2Ba8xOPoWWEGwFO3FBsOlzu-R1v0",
      "title": "AI-Driven Sales Automation -- The time is now!",
      "author": "",
      "published_date": "Jul 8, 2024"
    },
    {
      "link": "https://www.linkedin.com/pulse/dramatic-roi-ai-sales-process-automation-dmitry-shapiro-ds9tc",
      "image": "https://media.licdn.com/dms/image/v2/D5612AQGvabiginA5QA/article-cover_image-shrink_720_1280/article-cover_image-shrink_720_1280/0/1720285865490?e=2147483647&v=beta&t=LJLoaDtSNkv8E1pmgPo-SlfrHapYbOf5hgil5psbx7Y",
      "title": "Dramatic ROI -- AI Sales Process Automation",
      "author": "",
      "published_date": "Jul 6, 2024"
    },
    {
      "link": "https://www.linkedin.com/pulse/youve-never-seen-before-evolving-software-dmitry-shapiro",
      "image": "https://static.licdn.com/scds/common/u/img/pic/pic_pulse_stock_article_17.jpg",
      "title": "You've never seen this before - evolving software",
      "author": "",
      "published_date": "Sep 16, 2020"
    }
  ],
  "description": {},
  "activities": [
    {
      "link": "https://www.linkedin.com/posts/marquis-sampson_salestips-businessintelligence-companyresearch-activity-7350954619461627905-OkHc",
      "image": "https://media.licdn.com/dms/image/v2/D5605AQHnUxZNQf4nBQ/videocover-high/B56ZgPWGL_HMCI-/0/1752604135536?e=2147483647&v=beta&t=a3RCVDehSZ-YiDn3xClq9lnkblvUeLBeHvxcpc_tnYw",
      "title": "Tired of spending hours researching companies before a meeting? 🕵️‍♂️ Introducing the Cheat Sheet Generator! Your new secret weapon for instant…",
      "activity": "Liked by Dmitry Shapiro"
    }
  ],
  "volunteering": [],
  "certification": [],
  "people_also_viewed": [
    {
      "link": "https://www.linkedin.com/in/hyams?trk=public_profile_browsemap-profile",
      "name": "Chris Hyams",
      "summary": "",
      "location": "Austin, Texas Metropolitan Area"
    },
    {
      "link": "https://www.linkedin.com/in/jonvferrara?trk=public_profile_browsemap-profile",
      "name": "Jon Ferrara",
      "summary": "",
      "location": "Asheville, NC"
    },
    {
      "link": "https://www.linkedin.com/in/hillferguson?trk=public_profile_browsemap-profile",
      "name": "Hill Ferguson",
      "summary": "",
      "location": "San Francisco, CA"
    },
    {
      "link": "https://www.linkedin.com/in/vikas?trk=public_profile_browsemap-profile",
      "name": "Vikas Sharan",
      "summary": "",
      "location": "San Francisco Bay Area"
    },
    {
      "link": "https://www.linkedin.com/in/richardreed1992?trk=public_profile_browsemap-profile",
      "name": "Richard Reed",
      "summary": "",
      "location": "Las Vegas, NV"
    },
    {
      "link": "https://www.linkedin.com/in/mkorsunsky?trk=public_profile_browsemap-profile",
      "name": "Michael Korsunsky",
      "summary": "",
      "location": "Charlotte Metro"
    },
    {
      "link": "https://www.linkedin.com/in/rahulchaturvedimd?trk=public_profile_browsemap-profile",
      "name": "Rahul Chaturvedi",
      "summary": "CEO @ Moolex LLC | Start-ups, R&D, Project Capital, Business Strategy",
      "location": "New York, NY"
    },
    {
      "link": "https://www.linkedin.com/in/foxbrian?trk=public_profile_browsemap-profile",
      "name": "Brian Fox",
      "summary": "",
      "location": "Denver, CO"
    },
    {
      "link": "https://www.linkedin.com/in/drodio?trk=public_profile_browsemap-profile",
      "name": "DROdio - Daniel R. Odio",
      "summary": "",
      "location": "San Francisco Bay Area"
    },
    {
      "link": "https://www.linkedin.com/in/brett-mallory?trk=public_profile_browsemap-profile",
      "name": "Brett Mallory",
      "summary": "",
      "location": "Boston, MA"
    },
    {
      "link": "https://www.linkedin.com/in/jager?trk=public_profile_browsemap-profile",
      "name": "Jager McConnell",
      "summary": "",
      "location": "San Francisco, CA"
    },
    {
      "link": "https://th.linkedin.com/in/jchenard?trk=public_profile_browsemap-profile",
      "name": "Jesse Chenard",
      "summary": "",
      "location": "Amphoe Ban Pong"
    },
    {
      "link": "https://www.linkedin.com/in/davepowell?trk=public_profile_browsemap-profile",
      "name": "Dave Powell",
      "summary": "",
      "location": "Austin, TX"
    },
    {
      "link": "https://www.linkedin.com/in/chris-rothstein-4047012?trk=public_profile_browsemap-profile",
      "name": "Chris Rothstein",
      "summary": "",
      "location": "San Diego, CA"
    },
    {
      "link": "https://il.linkedin.com/in/liadagmon?trk=public_profile_browsemap-profile",
      "name": "Liad Agmon",
      "summary": "",
      "location": "Tel Aviv District, Israel"
    },
    {
      "link": "https://www.linkedin.com/in/masolova?trk=public_profile_browsemap-profile",
      "name": "Elena M.",
      "summary": "",
      "location": "San Francisco Bay Area"
    },
    {
      "link": "https://www.linkedin.com/in/haroldlee?trk=public_profile_browsemap-profile",
      "name": "Harry Lee",
      "summary": "",
      "location": "San Francisco Bay Area"
    },
    {
      "link": "https://www.linkedin.com/in/gt123?trk=public_profile_browsemap-profile",
      "name": "Greg T.",
      "summary": "",
      "location": "Ann Arbor, MI"
    },
    {
      "link": "https://www.linkedin.com/in/mcobrien?trk=public_profile_browsemap-profile",
      "name": "Mike O'Brien",
      "summary": "",
      "location": "Franklin, TN"
    },
    {
      "link": "https://www.linkedin.com/in/sgupta85?trk=public_profile_browsemap-profile",
      "name": "Sunny Gupta",
      "summary": "",
      "location": "Kirkland, WA"
    }
  ],
  "similar_profiles": [
    {
      "link": "https://www.linkedin.com/in/drshapiro?trk=public_profile_samename-profile",
      "name": "Dmitry Shapiro",
      "summary": "",
      "location": "Greater Boston"
    },
    {
      "link": "https://www.linkedin.com/in/dmitryshapiro?trk=public_profile_samename-profile",
      "name": "Dmitry Shapiro",
      "summary": "Problem Solver at cbconstruction",
      "location": "San Francisco, CA"
    },
    {
      "link": "https://www.linkedin.com/in/dmitry-shapiro-8692771?trk=public_profile_samename-profile",
      "name": "Dmitry Shapiro",
      "summary": "",
      "location": "Highland Park, IL"
    },
    {
      "link": "https://www.linkedin.com/in/dmitry-shapiro-10214218?trk=public_profile_samename-profile",
      "name": "Dmitry Shapiro",
      "summary": "--",
      "location": "New York City Metropolitan Area"
    }
  ],
  "recommendations": [
    {
      "link": "https://www.linkedin.com/in/jcuzens?trk=public_profile_recommendations",
      "name": "Jarrod Cuzens",
      "summary": "“Not only is Dmitry one of the most creative and personable individuals I have had the pleasure to work with but he is also one of those people that I will always admire for his ability to come up with and sell these ideas. If I had to sum up Dmitry in one word it would certainly be visionary.”"
    },
    {
      "link": "https://www.linkedin.com/in/sunnygault?trk=public_profile_recommendations",
      "name": "Sunny Gault",
      "summary": "“Dmitry is one of the most unusual CEO's I've ever met- but that's what makes him so great. You feel completely comfortable talking to him and sharing your ideas- which creates an ideal working environment. When you've got a good idea, he commends you for it and gives you the tools you need to go out there and tackle the world!”"
    }
  ],
  "publications": [],
  "courses": [],
  "languages": [
    {
      "name": "Russian",
      "level": "Limited working proficiency"
    },
    {
      "name": "English",
      "level": "Native or bilingual proficiency"
    }
  ],
  "organizations": [],
  "projects": [],
  "awards": [],
  "score": []
}

{
  "company_name": "MindStudio",
  "universal_name_id": "mindstudioai",
  "background_cover_image_url": "https://media.licdn.com/dms/image/v2/D563DAQHsfMJ3A0WRjQ/image-scale_191_1128/image-scale_191_1128/0/1682953888759/youai_ai_cover?e=2147483647&v=beta&t=OQmyNF7EJXFK4-FKvEgxmy7zn-g4P0MsCLubu4TzP_s",
  "linkedin_internal_id": "91188227",
  "profile_photo": "https://media.licdn.com/dms/image/v2/D560BAQHJWyoFxDx2vQ/company-logo_200_200/company-logo_200_200/0/1704388316703/youai_ai_logo?e=2147483647&v=beta&t=2u4H4OT-wjxp9NB44WQVO4AaE429XmlLjBGIKUJLOSs",
  "industry": "Technology, Information and Internet",
  "location": "",
  "tagline": "The fastest way to build AI-powered apps for your team.",
  "company_size_on_linkedin": "Discover all 29",
  "about": "Rapidly build custom AI applications and automations — no coding required.\nEasily mix and match the latest models from OpenAI, Anthropic, Google, Mistral, Meta, and more.",
  "website": "https://mindstudio.ai",
  "industries": "Technology, Information and Internet",
  "company_size": "11-50 employees",
  "headquarters": "",
  "type": "Privately Held",
  "founded": "2023",
  "specialties": "",
  "description": {},
  "locations": [],
  "employees": [
    {
      "employee_photo": "https://media.licdn.com/dms/image/v2/D5603AQHhQ4-UhGgpqA/profile-displayphoto-shrink_100_100/profile-displayphoto-shrink_100_100/0/1708916323677?e=2147483647&v=beta&t=8f09N2sXi6CpaIOvlnEoeGdAidR2gWrOsC1wZLjbC6E",
      "employee_name": "Dmitry Shapiro",
      "employee_position": "",
      "employee_profile_url": "https://www.linkedin.com/in/dmitry-shapiro-a2b1?trk=org-employees"
    },
    {
      "employee_photo": "https://media.licdn.com/dms/image/v2/D4E03AQHFqAOAyv6d1g/profile-displayphoto-shrink_100_100/B4EZbakNBNGkAU-/0/1747423646207?e=2147483647&v=beta&t=uqATJggybaqd-MsnJoDxM2ZPnmZTFSRXmeUUHMAboKo",
      "employee_name": "Aleksandar Pokrajac",
      "employee_position": "AI Innovation PM | Custom AI Solutions",
      "employee_profile_url": "https://rs.linkedin.com/in/pokrajac?trk=org-employees"
    },
    {
      "employee_photo": "https://media.licdn.com/dms/image/v2/D4D03AQEoOkPm9qMz4g/profile-displayphoto-scale_100_100/B4DZfFY.REH4Ag-/0/1751363348177?e=2147483647&v=beta&t=ZE49EIlCvWMKIn3EbImELcktUu8MSCKIGTD940x4Obc",
      "employee_name": "Gary Seath",
      "employee_position": "I help business people to connect with eachother. Who would you like to connect with?",
      "employee_profile_url": "https://za.linkedin.com/in/gary-seath?trk=org-employees"
    },
    {
      "employee_photo": "https://media.licdn.com/dms/image/v2/D4D03AQGUNXSrPOMSRQ/profile-displayphoto-shrink_100_100/B4DZU0NfqGG8AU-/0/1740337709318?e=2147483647&v=beta&t=7PSR7s8uGKlJ5z3Kgx251rPbIEcJBL9mxF1W-pIJgyk",
      "employee_name": "Jonathan M K.",
      "employee_position": "Head of GTM Growth Momentum | Founder GTM AI Academy & Cofounder AI Business Network | Business impact > Learning Tools | Proud Dad of Twins",
      "employee_profile_url": "https://www.linkedin.com/in/jmkmba?trk=org-employees"
    }
  ],
  "updates": [
    {
      "text": "New Fetch Slack Channel History block makes it easy to pull messages from a Slack channel into a MindStudio AI Agent and generate reports, analyze them, create summaries, etc.\n\nAvailable today, check it out!",
      "article_posted_date": "5d",
      "total_likes": "25",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "MindStudio University is live! 🎓\n\nIn MindStudio University, you'll find over 6 hours of step-by-step, in-depth tutorials on building AI agents. Hundreds of people have already levelled up their AI skills with our video courses. \n\nBookmark the link below!",
      "article_posted_date": "2w",
      "total_likes": "48",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "You could spend thousands of dollars testing ad copy... or you could use this agent to find the winning formula. \n\nExpert marketer Justin Brooke walks through it below.\nHow I built an AI agent that predicts winning ad copy BEFORE spending a dime...\n\nIf you could predict lottery numbers... would you?\n\nWell then, if you could predict winning ads... would you?\n\nLet me explain how it works...\n\nFirst, we trained an agent on all our customer data. We give our ad copy (or blog post, whatever you want) to the first agent. Which gives us feedback on the ad creative based on all our customer data.\n\nIt's so real, that it feels like I'm reading an actual customer support ticket from a real customer.\n\nThe feedback alone is amazing to learn from if you're not a great writer yet.\n\nBut then we take the original ad creative + feedback, and we send that to another agent that is trained on world class direct response copywriting tactics.\n\nIt then writes use 3 new variations for our original ad creative, based on our customers feedback. This is just like they used to do with focus groups and customer development calls - but with ai and automated.\n\nIf all it did was pump out expertly written variations (based on actual customer data) this would be worth gold, but we go another step.\n\nWe want to know which variation is statistically likely to outperform the others, so we push the feedback + variations to a data analyst agent to assign scores.\n\nScoring is based on how well the ad fits our customer data, the feedback from our customer simulator agent, and all known best practices for advertising.\n\nFinally, the data analyst agent pushes the Success Probability Score for each variation and the winning variation to the screen for the user to deploy.\n\nVOILA... Now your cousin Marv can write ads for you and with this machine it'll improve it based on actual market data, get 3 revisions by an expert, and then predict which revision is most likely to win.\n\nIf that's not worth millions, I don't know what is!\n\nIf you want to see the prompts and structure of how I built this, just comment \"agent\" below, and I'll DM you a loom video of how it works. MUST BE CONNECTED for me to be able to DM you.",
      "article_posted_date": "2w",
      "total_likes": "5",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "Build the tools that make your team call you an \"AI savant\" like David Cohn. \n\nKudos to Advance Local for being at the forefront of AI in journalism. They use nearly 300 MindStudio agents, like this one, in their newsrooms. An AI Tool I built is getting some praise/love from the Editor/VP of content at cleveland.com / The Plain Dealer. It's called \"Editor's Eyes\" and they are now running all their stories through it before publishing. \n\nhttps://lnkd.in/gx7wa-ng\n\n\"Starting this month, anyone in our newsroom who writes – including me -- must run their work through a tool we’ve developed to clean it up. It’s a tool called Editor’s Eyes, developed in-house by David Cohn, the AI savant for our parent company, Advance Local. David has taken our breath away with the number of tools he has developed for us this year.\nWriters need only paste their work into the tool, and seconds later they get a list of fixes they can make, with advice for making it read better. In effect, it gives every writer their own personal copy desk.\"\n\nA big shoutout to Christopher Quinn who embraces the changing landscape like a real leader. And to all the colleagues at Cleveland (and Advance throughout). You all keep me motivated and pumped for more.",
      "article_posted_date": "2w",
      "total_likes": "4",
      "article_title": "Cleveland OH Local News, Breaking News, Sports & Weather",
      "article_sub_title": "cleveland.com",
      "article_link": "https://www.linkedin.com/redir/redirect?url=https%3A%2F%2Fwww%2Ecleveland%2Ecom&urlhash=XJDJ&trk=organization_guest_main-feed-card_reshare_feed-article-content",
      "article_image": "https://media.licdn.com/dms/image/sync/v2/D5627AQFOr8xR0NqxvA/articleshare-shrink_1280_800/B56ZdeRUriGoAQ-/0/1749633289341?e=2147483647&v=beta&t=-iv_bcxztTop5Xe0-FHWc9YgRxYyHDw4M7D940QsXng"
    },
    {
      "text": "📸 𝗜𝗺𝗮𝗴𝗲𝗻 𝟰 𝗶𝘀 𝗻𝗼𝘄 𝗮𝘃𝗮𝗶𝗹𝗮𝗯𝗹𝗲 𝗼𝗻 𝗠𝗶𝗻𝗱𝗦𝘁𝘂𝗱𝗶𝗼!\nWe're thrilled to announce the launch of 𝗚𝗼𝗼𝗴𝗹𝗲'𝘀 𝗜𝗺𝗮𝗴𝗲𝗻 𝟰—the latest and most advanced text-to-image model—𝗻𝗼𝘄 𝗹𝗶𝘃𝗲 𝗼𝗻 𝗠𝗶𝗻𝗱𝗦𝘁𝘂𝗱𝗶𝗼.\n\nWhether you're an artist, a product designer, a marketer, or just curious, you can now generate 𝗽𝗵𝗼𝘁𝗼𝗿𝗲𝗮𝗹𝗶𝘀𝘁𝗶𝗰, 𝗵𝗶𝗴𝗵-𝗿𝗲𝘀𝗼𝗹𝘂𝘁𝗶𝗼𝗻 𝘃𝗶𝘀𝘂𝗮𝗹𝘀 from text with unprecedented fidelity—directly inside your AI workflows. \n\n✨ 𝗪𝗵𝘆 𝗜𝗺𝗮𝗴𝗲𝗻 𝟰?\n\n🖼️ - 𝗣𝗵𝗼𝘁𝗼-𝗿𝗲𝗮𝗹𝗶𝘀𝗺 that rivals actual photography\n🔍 - 𝗦𝗵𝗮𝗿𝗽𝗲𝗿 𝗰𝗹𝗮𝗿𝗶𝘁𝘆 and 𝗶𝗺𝗽𝗿𝗼𝘃𝗲𝗱 𝘁𝗲𝘅𝘁 𝗿𝗲𝗻𝗱𝗲𝗿𝗶𝗻𝗴\n🔠 - 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝘀𝗽𝗲𝗹𝗹𝗶𝗻𝗴 𝗮𝗻𝗱 𝘁𝘆𝗽𝗼𝗴𝗿𝗮𝗽𝗵𝘆 for comics, packaging, and design\n🎨 - 𝗦𝘂𝗽𝗽𝗼𝗿𝘁𝘀 𝗱𝗶𝘃𝗲𝗿𝘀𝗲 𝘀𝘁𝘆𝗹𝗲𝘀: from macro wildlife photography to vintage illustrations, impressionism, 3D renders, and more\n\n🌟  𝗧𝗿𝘆 𝗶𝘁 𝗻𝗼𝘄: mindstudio.ai → Create → Generate Image\n\n Let your imagination do the talking. We'll take care of the rendering.",
      "article_posted_date": "2w",
      "total_likes": "29",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "Students, recent grads & dropouts: the #1 skill you can learn right now → building AI agents. \n\nYour school isn't teaching the skills companies are desperate for... but we are. \n\nJoin the next cohort with our CEO, Dmitry Shapiro, to build your AI portfolio and connect with employers who are looking for talent.College students, recent college graduates, parents of college students...\n\nI am teaching a free 2 week course -- Learn to build AI agents — the most in-demand skill today!\n\nHere’s how it works:\n\nDuration: 2 weeks\nFormat: Virtual (Zoom-based training + self-paced projects)\nTime commitment: ~10 hours (a few 1-hour Zoom sessions + project work)\nTools: You’ll use MindStudio to build real AI agents\nCost: Completely free\nSpots: Limited (20–40 per cohort)\n\nMore details in comments 👇\n\nWho else should know about this? @ mention them.",
      "article_posted_date": "3w",
      "total_likes": "4",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "After leading tech teams at Walmart, Yahoo, and Strava, Claude Jones and the others behind the San Diego Tech Hub are launching do.it.withAI, a community-based platform to train AI builders. \n\nTheir first AI bootcamp is free, and powered by MindStudio. \n\nApply below!👋 Hey Community! \n\nI’m back… and ready to make an impact.\n\nAfter being let go from Centr, I finally had space to reflect. Not just on what I wanted to do next, but on what people around me were struggling with.\n\nIn every conversation with friends, colleagues, and community members, one theme kept surfacing: AI is already transforming how we live and work, and for many, it's creating real fear.\n\nNot just fear of being replaced, but fear of being left behind.  For many, it’s not even about job loss. It’s about not knowing where to start or how to keep up.\nThat stuck with me.  Because I’ve felt it too.\n\nWhat people need isn’t just access to AI tools.  They need agency — the power to learn, build, and adapt on their own terms.\n\nSo I got together with some San Diego Tech Hub conduits, and we created do.it.withAI, a community-driven platform built to train the next generation of AI Agent Builders.\n\nWe’ve partnered with Dmitry Shapiro at MindStudio to launch a free live pilot AI Bootcamp, limited to 100 participants.\n\nThis isn’t a giveaway. It’s a co-build. Your experience and feedback will help shape how we train the next generation of AI Agent Builders.\n\n🎥 Check out the video below to see what we’ve been building. 👇 \n\nOh, before I go, two quick points: \n\n1. If this resonated, share it. Someone in your network might need it.\n2. If you’re ready to build or brainstorm, DM me. Let’s make moves.\n\n#AI #Community #TechForGood #DoItWithAI #AgencyOverAccess #AIAgents\n\nNotes: site built with Lovable, Supabase, ChatGPT, and Human creativity in the loop.",
      "article_posted_date": "3w",
      "total_likes": "5",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "The UK Government is going deep on AI—and MindStudio is right at the center. \n\nYesterday, we led a hands-on session with the Incubator for Artificial Intelligence, an agile team inside the government, tasked with scaling AI tools across departments. Their mission: use AI to drive public good.\n\nIn an hour, they built working AI agents with MindStudio. Real prototypes for real public sector problems.\n\nThis is the playbook for modern governments:\n\n🛠️ Lightweight tools – skip the external contracts\n🔑 Broad access – let teams solve their own problems\n⚡ Fast results – prototypes in hours, not months\n\nMindStudio makes it happen.",
      "article_posted_date": "1mo",
      "total_likes": "22",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "Now available: Use Self-Hosted Models in your AI agents 🧠\n\nYou can now connect your own self-hosted AI models — whether they’re running locally on your device or hosted in the cloud. Works for any OpenAI-compatible model.\n\nMindStudio already gives you access to 150+ models out of the box — Claude 4, GPT-4o, and more — no API keys required.\n\nNow you can go beyond the flagships and bring your own.\n\n🛠️ Just tinkering? It’s included in MindStudio Pro for personal use.\n🏢 Deploying for business? Get in touch with us about Enterprise.",
      "article_posted_date": "1mo",
      "total_likes": "23",
      "article_title": "",
      "article_sub_title": ""
    },
    {
      "text": "Just shipped: 11 new blocks for transforming text in MindStudio 🔁\n\nEverything you actually need to do with text — now built in.\nNo prompt required. Just drop the block in and go.\n\nHere’s what’s new:\n\n📝 Summarize Text – Get the TL;DR\n\n🧠 Improve Writing – Fix spelling, grammar, clarity\n\n🌍 Translate Text – Any language, instantly\n\n👥 Rewrite for Audience – Adapt for different personas\n\n🎓 Convert Reading Level – Make it simpler or more advanced\n\n🔀 Merge Text – Combine multiple inputs into one\n\n💬 Adjust Tone – Casual, formal, fun — your call\n\n✅ Align to Brand – Stay on voice, every time\n\n✏️ Rewrite Text – Give it a rule, and it’ll rewrite at scale\n\n🧱 Convert to Format – Repurpose content into a blog, email, post — whatever you need\n\n🧹 Extract from Text – Cleanly pull structured info from messy input\n\nLive now on MindStudio.",
      "article_posted_date": "1mo\n  \n                      \n                      \n                        Edited",
      "total_likes": "14",
      "article_title": "",
      "article_sub_title": ""
    }
  ],
  "similar_companies": [
    {
      "link": "https://nl.linkedin.com/company/jaide-ai?trk=similar-pages",
      "name": "Jaide AI",
      "summary": "Software Development",
      "location": "Reeuwijk, Zuid holland"
    },
    {
      "link": "https://www.linkedin.com/company/syntheticusers?trk=similar-pages",
      "name": "Synthetic Users",
      "summary": "Technology, Information and Internet",
      "location": "Technology, Information and Internet"
    },
    {
      "link": "https://www.linkedin.com/company/withkoji?trk=similar-pages",
      "name": "Koji",
      "summary": "Technology, Information and Internet",
      "location": "San Diego, CA"
    },
    {
      "link": "https://cz.linkedin.com/company/itsmakehq?trk=similar-pages",
      "name": "Make",
      "summary": "Software Development",
      "location": "Prague, Praha 8"
    },
    {
      "link": "https://www.linkedin.com/company/perplexity-ai?trk=similar-pages",
      "name": "Perplexity",
      "summary": "Software Development",
      "location": "San Francisco, California"
    },
    {
      "link": "https://www.linkedin.com/company/gamma-app?trk=similar-pages",
      "name": "Gamma",
      "summary": "Software Development",
      "location": "Software Development"
    },
    {
      "link": "https://dk.linkedin.com/company/uizard?trk=similar-pages",
      "name": "Uizard by Miro Labs",
      "summary": "Software Development",
      "location": "Software Development"
    },
    {
      "link": "https://www.linkedin.com/company/flowiseai?trk=similar-pages",
      "name": "FlowiseAI (YC S23)",
      "summary": "Software Development",
      "location": "San Francisco, California"
    },
    {
      "link": "https://www.linkedin.com/company/hunchtools?trk=similar-pages",
      "name": "Hunch",
      "summary": "Technology, Information and Internet",
      "location": "San Francisco, California"
    },
    {
      "link": "https://www.linkedin.com/company/workera-ai?trk=similar-pages",
      "name": "Workera",
      "summary": "Technology, Information and Internet",
      "location": "Palo Alto, California"
    }
  ],
  "affiliated_companies": [],
  "product": []
}

Writing Prompts

Learn about prompt engineering for AI tasks.

Prompts are the foundation of how your AI Agent understands and executes tasks. A well-written prompt ensures your AI Agent delivers precise and meaningful results.

What is a Prompt?

A prompt is a set of instructions that tells the AI Agent what to do. It provides the context and parameters for generating the desired output. Just like giving someone directions, a good prompt ensures the AI knows exactly what to deliver. In MindStudio, prompts can be used at two levels: System Prompts and Block Prompts.

System Prompt

The System Prompt appears in the Prompt Tab of a workflow file. It serves as the AI Agent's core instructions, defining its role, capabilities, and constraints and acts as the foundation, guiding how the AI behaves throughout the workflow.

When you write a system prompt, you’re establishing the AI's "role" and general approach to tasks. Every action in the workflow will follow this overarching guidance unless overridden by specific block prompts.

You can write a system prompt manually by typing into the blank space of the Prompt Tab, or you can click on the Generate Prompt button at the bottom left to have the Prompt Generator to write the prompt for you.

Using the Prompt Generator

Click on the Generate Prompt button at the bottom left of the prompt area. This will open the Prompt Generator Modal.
Using natural language, enter a brief description of what your AI Agent is supposed to do.
1. Example:
After entering your description, click the Generate button. The system will automatically create a structured prompt based on your input.

Block Prompt

A block prompt is used for specific tasks within the workflow. While the system prompt provides overall guidance, a block prompt gives detailed instructions for a particular step.

Example of a Block Prompt (within a Generate Text block):

Here, the block prompt tells the AI to focus on summarizing a specific piece of information (e.g., the result of a Google search). This allows you to customize how the AI performs at different stages of the workflow.

How System Prompts and Block Prompts Work Together

System Prompt

Block Prompt

Key Aspects of Prompt Writing

A good prompt is clear, specific, and provides enough context for the AI to understand what’s needed.

Key aspects of an effective Prompts:

Clarity: Use straightforward language to avoid confusion. Clearly define the AI Agent's purpose and main tasks
Specificity: Be detailed about what you want. Specify any limitations or guidelines that your AI Agent should follow when executing its task
Context: Explain the purpose or provide a scenario to guide the AI. Context can be provided via direct instruction or by calling Variables within a prompt.

Examples of Good Prompts vs. Bad Prompts

Scenario

Good Prompt

Bad Prompt

Using Markdown to Write Prompts

Markdown is a lightweight formatting language that can be used to structure prompts, making them clearer and easier for the AI to interpret. By using Markdown, you can organize your instructions in a readable and visually structured way.

Why Use Markdown in Prompts?

Clarity: Break down complex instructions into digestible sections.
Readability: Using formatting like headers, bullet points, and code blocks to make prompts easier to follow.
Consistency: Ensure your prompt has a standardized structure, especially for complex tasks.
Debugging: Allows for easier debugging and refinement of prompts.

Common Markdown Elements for Prompts

Here are some common Markdown features and how to use them:

Headers
- Purpose: Use headers to introduce sections or tasks.
- Variations:
  - #

Example System Prompt Using Markdown

For more information on Markdown, see the .

Auto-Enhance Prompts

The Auto-Enhance feature helps polish and refine your prompts after you've made modifications. It analyzes your prompt for clarity, formatting, and potential ambiguities, then suggests improvements while maintaining your original intent.

How Auto-Enhance Works

Access the feature: Click the "Enhance" button in the bottom bar or use the keyboard shortcut Option + K
Review suggestions: The system presents an enhanced version of your prompt with improvements in:
- Grammar and spelling
- Markdown formatting consistency

This feature works both in the main Prompt Tab and in fullscreen Send Message blocks, making it easy to maintain high-quality prompts throughout your workflow.

Pro Tip: Auto-Enhance is particularly useful when you've started with a generated prompt and made manual modifications, helping ensure your customizations maintain professional quality and clarity.

Creating a structured template lets the model knows exactly where and how to inject content. By pairing a template with your data, you get consistent, predictable results every time.

Writing Templates

A template is a block of text (in Markdown, HTML, or JSON) containing placeholders for your variables. When your workflow runs, MindStudio replaces those placeholders with the actual values you’ve collected (via user inputs, scraped data, files, etc.) and sends the combined prompt to the model.

Placeholders can use either XML-style tags <company>…</company> or square-bracket notation [company].
Inside a Generate Text or Generate Asset block, you supply:
1. Your data (e.g., user inputs saved to variables)

Why Use Templates?

Consistency: Ensure every output follows the same format.
Clarity: Guide the model step-by-step—no guessing about headings, sections, or order.
Scalability: One template handles countless inputs.
Flexibility: Swap templates for different formats (HTML vs. Markdown vs. JSON) without rewriting your workflow.

Writing Templates

Using Square Bracket Notation

Square brackets are ideal for quick and easy placeholders in Markdown templates or ad-hoc prompts. They are simple to identify in plain text and work best when you have single-word placeholders like [title] or [body].

Use them when writing short, straightforward templates, such as bullet lists or paragraphs, where the format resembles a "fill-in-the-blanks" approach.

Example prompt that uses Square brackets:

Using XML Tags

XML-style tags clearly show where each placeholder starts and ends to help avoid confusion with nearby text or symbols. This is particularly helpful for making sure variables are clearly displayed to the AI model.

Prompt Example (Generate Blog Post):

You can also use them when you have longer placeholders, or for creating HTML templates with the Generate Asset block.

HTML Example (Marketing Email):

For more information on creating HTML templates please review the article.

Best Practices for Template Writing

Name your placeholders clearly (e.g. [project_summary], not [x]).
Wrap variables in tags to avoid ambiguity <budget>{{budgetVariable}}<budget> .
Test incrementally: start with one section, preview, then expand.

AI Agents for HR

Resume Screening for Key Qualifications

Use an AI Agent to screen resumes for key qualifications, identifying the best candidates faster.

Import candidate resumes and job description criteria.
Schedule AI Agent to run upon resume submission.
Use AI to analyze resumes for matching qualifications and relevant experience.
Output a shortlist of qualified candidates to the recruitment dashboard.
Notify recruiters with the top candidates for review.

Candidate Ranking by Job Fit

Use an AI Agent to rank candidates by job fit, streamlining focus on top prospects.

Import candidate profiles, resumes, and job requirements.
Schedule AI Agent to run during the candidate evaluation phase.
Use AI to score and rank candidates based on job fit criteria (e.g., skills, experience, cultural alignment).
Output ranked candidate lists to the recruitment system.

Inclusive Job Description Optimization

Use an AI Agent to improve job descriptions, boosting inclusivity and appeal.

Import draft job descriptions.
Schedule AI Agent to run during job posting preparation.
Use AI to analyze language for inclusivity and suggest improvements.
Output revised job descriptions with inclusivity enhancements.

Candidate Engagement Scoring

Use an AI Agent to score candidate engagement with job postings, emails, and interviews.

Import engagement data (e.g., application interactions, email responses, interview feedback).
Schedule AI Agent to run after key engagement milestones.
Use AI to calculate an engagement score for each candidate.
Output engagement scores to the recruitment dashboard.

Employee Role Matching

Use an AI Agent to match employees to new roles, considering skills and performance data.

Import employee skills, performance metrics, and open role criteria.
Schedule AI Agent to run during role transition planning.
Use AI to match employees to roles based on skills and suitability.
Output role recommendations to the project management system.

Bias Detection in Hiring Processes

Use an AI Agent to check hiring processes and job descriptions for potential biases.

Import job descriptions, interview scripts, and hiring process data.
Schedule AI Agent to run before launching new hiring initiatives.
Use AI to identify and flag potential biases in language or processes.
Output recommendations to improve inclusivity and reduce bias.

Offer Letter Generation

Use an AI Agent to generate offer letters, incorporating relevant candidate and policy details.

Import candidate details and HR policy templates.
Schedule AI Agent to run upon candidate selection for an offer.
Use AI to generate customized offer letters with appropriate terms.
Output finalized offer letters for review.

Post-Interview Feedback Analysis

Use an AI Agent to analyze post-interview feedback, enhancing the hiring process.

Import interview feedback data (e.g., notes, ratings, comments).
Schedule AI Agent to run after interview completion.
Use AI to identify trends, strengths, and areas for improvement.
Output actionable insights to the recruiting team.

Tailored Interview Question Suggestions

Use an AI Agent to suggest interview questions aligned with resumes and role requirements.

Import candidate resumes and job descriptions.
Schedule AI Agent to run during interview preparation.
Use AI to generate tailored interview questions based on candidate profiles and role needs.
Output suggested questions to the interview scheduling system.

AI Agents for Marketing

Personalized Website and Email Content

Use an AI Agent to personalize website and email content, enhancing user engagement.

Import user behavior and demographic data.
Schedule AI Agent to run in real-time for website interactions or prior to email sends.
Use AI to generate personalized content based on user preferences.
Output tailored content to the website CMS or email platform.
Notify the marketing team with engagement performance metrics.

Ad Copy Variations for A/B Testing

Use an AI Agent to generate multiple ad copy variations for A/B testing across platforms.

Import campaign goals, target audience details, and base messaging.
Schedule AI Agent to run during ad copy planning.
Use AI to create multiple ad copy variations tailored to the audience.
Output variations to the ad platform for A/B testing.

Audience Segmentation for Campaigns

Use an AI Agent to segment audiences by behavior and demographics, tailoring campaigns for greater impact.

Import audience data (e.g., website behavior, purchase history, demographics).
Schedule AI Agent to run before campaign creation.
Use AI to identify and segment audiences based on patterns and preferences.
Output audience segments to the marketing platform.

Use an AI Agent to break down long-form content into digestible pieces for social platforms.

Import long-form content (e.g., articles, videos, reports).
Schedule AI Agent to run during content planning.
Use AI to generate shorter, platform-optimized content snippets.
Output snippets to the social media scheduler or asset library.

Visual Asset Categorization

Use an AI Agent to categorize visual assets, making images and videos easier to find and repurpose.

Import visual asset metadata and file library.
Schedule AI Agent to run periodically for new uploads.
Use AI to categorize assets by type, theme, and relevance.
Output categorized assets to the media library system.

Use an AI Agent to monitor social media trends, identifying emerging topics and keywords.

Import social media data streams (e.g., hashtags, posts, mentions).
Schedule AI Agent to run continuously or at regular intervals.
Use AI to analyze trends and identify emerging topics.
Output trend insights to the research dashboard.

Email Send Time Optimization

Use an AI Agent to optimize email send times, maximizing engagement per individual subscriber.

Import subscriber engagement data (e.g., open rates, click rates, time zones).
Schedule AI Agent to run before email campaign launches.
Use AI to determine optimal send times for each subscriber.
Output send-time recommendations to the email platform.

Campaign Success Prediction

Use an AI Agent to predict campaign success based on past data, adjusting strategies before launch.

Import past campaign data (e.g., performance metrics, audience engagement).
Schedule AI Agent to run during campaign planning.
Use AI to predict success probabilities and identify potential adjustments.
Output predictions and recommendations to the campaign planning tool.

Use an AI Agent to analyze social media sentiment, assessing public response to promotions.

Import social media mentions and comments data.
Schedule AI Agent to run during or after campaign launches.
Use AI to analyze sentiment trends (positive, neutral, negative).
Output sentiment insights to the brand team’s analytics dashboard.

AI Agents for Customer Service

Inquiry Categorization and Routing

Use an AI Agent to categorize inquiries, ensuring they are routed to the right agents.

Import inquiry data (e.g., email content, chat transcripts, call summaries).
Schedule AI Agent to run in real-time as inquiries arrive.
Use AI to classify inquiries by type and urgency.
Output categorized inquiries to the ticketing or routing system.
Notify agents with assigned tickets.

Pre-Written Response Suggestions

Use an AI Agent to suggest pre-written responses, enabling faster and more accurate replies.

Import past inquiry and response data.
Schedule AI Agent to run during live chat or email response drafting.
Use AI to recommend pre-written responses based on inquiry context.
Output response suggestions to the agent interface.

Interaction Summarization for Support Agents

Use an AI Agent to summarize past customer interactions, giving agents valuable context for responses.

Import interaction histories (e.g., tickets, chats, calls).
Schedule AI Agent to run when a new ticket or chat is opened.
Use AI to generate a concise summary of past interactions.
Output the summary to the agent's CRM view.

Real-Time Multilingual Support

Use an AI Agent for real-time translation, allowing agents to assist customers in various languages.

Import incoming messages or transcripts requiring translation.
Schedule AI Agent to run in real-time during customer interactions.
Use AI to translate messages into the agent’s preferred language and vice versa.
Output translated text to the chat or ticketing interface.

Ticket Prioritization for Urgent Issues

Use an AI Agent to prioritize tickets, ensuring urgent issues are handled promptly.

Import ticket details (e.g., content, metadata, timestamps).
Schedule AI Agent to run as tickets are created or updated.
Use AI to assign priority scores based on urgency criteria.
Output prioritized tickets to the support queue.

Complex Case Escalation

Use an AI Agent to identify and escalate complex cases to the support escalation team.

Import case details (e.g., ticket content, resolution attempts).
Schedule AI Agent to run after initial support actions fail.
Use AI to flag cases with high complexity or repeated issues.
Output flagged cases to the escalation queue.

Customer Satisfaction Prediction

Use an AI Agent to predict customer satisfaction, enabling proactive improvements before cases close.

Import ongoing case data (e.g., response times, sentiment analysis).
Schedule AI Agent to run periodically during case resolution.
Use AI to predict satisfaction levels based on historical trends.
Output satisfaction predictions and improvement recommendations.

Automated Data Entry for CRM

Use an AI Agent to automate data entry by extracting customer details from forms.

Import completed customer forms or inquiry details.
Schedule AI Agent to run upon form submission.
Use AI to extract and validate key customer information (e.g., name, email, issue type).
Output extracted data to the CRM fields.

Customer History Monitoring for QA

Use an AI Agent to monitor customer histories, identifying potential issues before they arise.

Import customer interaction histories and account data.
Schedule AI Agent to run periodically or before renewals or follow-ups.
Use AI to identify patterns indicating potential dissatisfaction or recurring issues.
Output flagged accounts to the QA or customer success team.

AI Agents for Finance

Automated Invoice Processing

Use an AI Agent to categorize and verify invoice data, enabling swift approvals.

Import invoice details (e.g., supplier, amounts, due dates).
Schedule AI Agent to run upon invoice submission.
Use AI to categorize invoices and verify data accuracy.
Output approved invoices to the accounting system.
Notify the accounting team with flagged issues for review.

Expense Categorization for Bookkeeping

Use an AI Agent to categorize expenses, improving accuracy and saving time for small businesses.

Import transaction data (e.g., receipts, bank statements).
Schedule AI Agent to run daily or weekly.
Use AI to categorize expenses based on predefined categories.
Output categorized data to the bookkeeping system.

Real-Time Fraud Detection

Use an AI Agent to flag suspicious transactions, enabling real-time fraud detection.

Import transaction data (e.g., account details, timestamps).
Schedule AI Agent to run continuously or in near real-time.
Use AI to identify patterns indicative of fraud.
Output flagged transactions to the fraud monitoring dashboard.

Transaction Classification for Tax Prep

Use an AI Agent to classify transactions, simplifying compliance and reducing tax prep time.

Import transaction records (e.g., sales, expenses, payments).
Schedule AI Agent to run quarterly or as needed.
Use AI to classify transactions by tax categories.
Output classified data to the tax preparation system.

Transaction Reconciliation

Use an AI Agent to match transactions across systems, identifying discrepancies for reconciliation.

Import transaction data from multiple financial systems.
Schedule AI Agent to run daily or weekly.
Use AI to match transactions and flag discrepancies.
Output reconciliation reports to the financial dashboard.

Payment Prioritization for Cash Flow Management

Use an AI Agent to suggest payment prioritization, balancing due dates and financial health.

Import payment obligations and cash flow data.
Schedule AI Agent to run during payment planning.
Use AI to prioritize payments based on due dates and cash availability.
Output payment recommendations to the finance team.

Revenue Forecasting for Budget Planning

Use an AI Agent to forecast revenue, enabling better financial planning for a non-profit.

Import historical revenue and donor data.
Schedule AI Agent to run quarterly or annually.
Use AI to predict revenue trends and donor behavior.
Output revenue forecasts to the budgeting tool.

Spending Pattern Analysis in Procurement

Use an AI Agent to detect unusual spending patterns, flagging anomalies in procurement.

Import procurement data (e.g., vendor, amounts, purchase categories).
Schedule AI Agent to run after new purchase orders.
Use AI to analyze patterns and identify anomalies.
Output flagged issues to the procurement dashboard.

Supplier Risk Assessment

Use an AI Agent to assess supplier risk, enabling data-driven procurement decisions.

Import supplier data (e.g., performance, payment history).
Schedule AI Agent to run periodically or during supplier reviews.
Use AI to evaluate supplier risk based on performance and trends.
Output risk scores to the supply chain finance system.