Make Your Own AI-Powered Research Assistant
TLDRThis video offers a step-by-step guide to creating a custom AI-powered research assistant that can automate industry-specific research at a user-defined time and format the findings for easy consumption via email, Slack, or social media. The process involves using three key tools: Make, an automation platform for connecting various services; Perplexity AI, an advanced search assistant; and the OpenAI Playground for creating custom GPT models. The tutorial covers setting up API connections, crafting effective prompts for Perplexity, and using custom GPTs to format research into digestible content for different platforms. Additionally, the video provides a troubleshooting guide for common issues and offers a ready-to-use bundle for those who prefer a pre-configured solution.
Takeaways
- 🤖 Utilize 'Make' to connect platforms without native integrations, like Chat GBT with Link D, Slack, or Medium.
- 🧠 Implement Perplexity AI for an advanced search assistant that can replace traditional search engines.
- 🚀 Access the playground of Chat GPT to create custom GPTs tailored to specific tasks.
- 💳 Purchase the connection from Perplexity to Make, which is necessary for the integration to work.
- 🔑 Secure an API key from Perplexity and add credits to your account for the search functionality.
- 📝 Create a detailed prompt for Perplexity to ensure the research output meets your requirements.
- 🔍 Test the prompt in Perplexity AI to verify the quality and relevance of the research results.
- ✉️ Set up an automated email to receive the research findings in a formatted and digestible manner.
- ⏰ Schedule the research assistant to run at specific times, such as every Monday at 6:00 a.m.
- 🔄 Use a router to direct the research output to various channels, including email, Slack, or social media profiles.
- 📈 For more digestible content, employ custom GPTs created in the playground to transform research into specific formats like LinkedIn posts.
Q & A
What is the purpose of the AI-powered research assistant discussed in the video?
-The AI-powered research assistant is designed to automate comprehensive research for a specific industry at a user-defined time. It can send the research findings in a well-formatted manner via email, Slack, or other communication channels, and even directly to social media profiles.
Which tools are essential for creating the custom AI-powered research assistant?
-The essential tools are Make (for connecting platforms without native connections), Perplexity AI (an AI-powered search assistant), and access to the playground of Chat GPT for creating custom GPTs.
How can one sign up for an account on Make?
-You can sign up for an account on Make using the link provided in the video description. It is an affiliate link, and using it helps the creators at no additional cost to the user.
What is the role of Perplexity AI in the research assistant?
-Perplexity AI is the key component that replaces Google search and performs the actual research based on the prompt given to it. It is used to gather and format the research data.
How does one access custom GPTs through the API?
-Custom GPTs can be accessed through the API by creating them in the playground of Open AI, not through the front end of Chat GPT.
What is the process of purchasing the connection from Perplexity to Make?
-After clicking on Perplexity in Make, you are taken to the integration page where you can purchase the connection for approximately $50. After the purchase, you receive an access code via email from the developers.
How does one obtain the API key from Perplexity?
-To obtain the API key, you need to go to your Perplexity account settings, access the API section, and insert some credits. The API key will be available for you to use in the Make connection.
What is the importance of crafting the right prompt for Perplexity?
-The prompt is crucial as it determines the quality and relevance of the research output from Perplexity. It needs to be detailed and specific to the user's research requirements.
How can the research assistant send the research directly to the user's email?
-After the research is completed by Perplexity, Make can be configured to send an email with the research content, which is formatted in HTML for better presentation.
What is the significance of setting a schedule for the research assistant?
-Scheduling ensures that the research assistant runs at a specific time, such as every Monday morning, to provide regular and timely updates without manual intervention.
How can the research assistant format the research into a LinkedIn post?
-By using a custom GPT created in the Open AI playground, which is trained to convert research into a LinkedIn post format, the assistant can generate a digestible and engaging post ready for sharing.
What is the benefit of using a router in the Make scenario?
-A router allows the research to be sent in different formats to various platforms, such as email, Slack, or directly to social media, providing flexibility in how and where the research is presented.
Outlines
🚀 Introduction to AI-Powered Research Assistant
This paragraph introduces the concept of an AI-powered research assistant that automates the research process for specific industries at chosen times. The assistant compiles comprehensive research and delivers it through preferred communication channels like email or Slack. It can even post directly to social media profiles. The key components mentioned are Make (a tool for connecting platforms without native connections), Perplexity (an AI-powered search assistant), and access to the OpenAI playground for creating custom GPT models. The paragraph also provides a link for signing up to Make and emphasizes its utility as a no-code automation tool.
🔍 Setting Up the Research Assistant with Perplexity
The second paragraph details the process of setting up the research assistant using Perplexity. It emphasizes the importance of creating the right prompt for Perplexity to ensure the research output meets the user's needs. The paragraph provides a template for the prompt and explains how to test it in Perplexity AI. It also guides on how to create a scenario in Make, connect it with Perplexity, and use the API key to establish the connection. The process includes purchasing a connection from Perplexity, installing it, and setting up the chat completion with the appropriate parameters.
✉️ Automating Email Delivery of Research Findings
This paragraph explains how to automate the delivery of research findings via email. It guides on setting up an action in Make to send an email with the research data formatted in HTML for better presentation. The process includes selecting the email action, setting the subject, and choosing the content from the Perplexity search output. It also covers troubleshooting steps if the email content does not appear correctly and emphasizes the importance of saving the scenario after each step.
📅 Scheduling and Further Customization of the Research Assistant
The fourth paragraph discusses scheduling the research assistant to run at specific times, such as every Monday morning, to provide regular updates. It also explores further customization options, like using custom GPT models to transform research into various formats like LinkedIn posts. The paragraph explains how to use Make's router to direct the research output to different channels and how to create custom GPT assistants in the OpenAI playground for specific tasks. It concludes with an offer to purchase a bundle that includes a JSON file with the research assistant setup, a prompt engineering tool, and other resources to simplify the process.
Mindmap
Keywords
💡AI-Powered Research Assistant
💡Make (Zapier-like tool)
💡Perplexity AI
💡API Keys
💡Custom GPT
💡Prompt Engineering
💡HTML
💡Slack
💡LinkedIn Post
💡JSON File
💡Stand Store
Highlights
Create a custom AI-powered research assistant to automate industry-specific research.
The assistant can send research findings via email, Slack, or directly to social media profiles.
Utilize Make.com to connect platforms without native integrations, such as ChatGBT, Link D, and Slack.
Perplexity AI is recommended as a search tool that has replaced Google search for the presenter.
Access to the OpenAI playground is required for creating custom GPT models.
Make.com offers a no-code automation tool for creating scenarios and connecting various services.
An affiliate link is provided for signing up on Make.com, which supports the presenter's channel.
A one-time payment of £37 (~$50) is required for the Perplexity AI connection on Make.com.
The presenter provides a template for creating an effective prompt for Perplexity AI.
Test the prompt in Perplexity AI to ensure it generates the desired research output.
Create a chat completion action in Make.com using the Perplexity AI connection.
Customize the message content for the Perplexity AI search and set the role as 'user'.
Save the action and run it to perform the search and obtain the data needed for the research assistant.
Set up an email action in Make.com to receive the research findings in a formatted email.
Choose the 'HTML' format for the email content to maintain formatting and readability.
Schedule the research assistant to run automatically at a specified time, such as every Monday at 6:00 AM.
Use custom GPT models in the OpenAI playground to transform research into various formats like LinkedIn posts.
Create a router in Make.com to direct the research output to different channels like email, Slack, or social media.
The presenter offers a bundle with a JSON file and additional resources to simplify the setup process for users.
The bundle includes a JSON file, a prompt engineer, a copywriter prompt, and the AR Power research assistant prompt.
The complete setup is available for purchase on the presenter's store for a nominal fee.