In the video "AI jako backend - ChatGPT do prototypowania?", Kuba Mrugalski demonstrates how to use ChatGPT as a backend for prototyping without coding by setting up a webhook to communicate with the GPT chatbot. He uses Make and Hopscotch tools to generate a webhook URL to send data to GPT via Open AI and retrieve responses in a JSON format. He then shows how to efficiently run the backend and choose the best fitting hashtags for the text input. Lastly, he offers a waitlist for a secret project on using OpenAI as a programmer for faster and more efficient prototyping.
00:00:00 In this section, Kuba Mrugalski explains how to generate a backend using artificial intelligence without code. He uses Chat GPT to return responses in a format useful for his application, which generates API that retrieves text input and returns hashtags adapted to that text. Kuba uses two tools to make this process easy: Make and Hopscotch. By using Make, he generates a webhook URL to which he sends data by HTTP. Open AI is used to send his information to GPT. Lastly, Kuba uses another webhook, webhook response, to retrieve the response from GPT.
00:05:00 In this section, the speaker demonstrates how to use ChatGPT as a backend for prototyping by setting up a webhook to communicate with the GPT chatbot. They explain how to retrieve and display the list of available responses and choose the best fitting hashtags. The speaker shows how to convert the response format from text to JSON and how to run the backend quickly and efficiently. They conclude by offering a link to a waiting list for a secret project on using OpenAI as a programmer for faster and more efficient prototyping.