Introduction

[00:00:00]
>> Hi, welcome to ChatGPT for Web Developers. We're going to cover today quickly, as a first look, how artificial intelligence and LLMs work for web developers. And we will talk about that later. So what's an LLM? So my name is Maximiliano Firtman. Let me go really quickly about myself.

[00:00:23]
I'm from Argentina, I live in Buenos Aires, and I'm a web developer and also a mobile app developer. I'm actually excited as most people around the world about AI, and also a little scared. It's time between both worlds. I have authored many books, a lot of courses on different platforms, and of course the best one, Frontend Masters.

[00:00:44]
You will find a lot of workshops and courses that I have been doing on mobile app development and also web development. But this is a different kind of worship. It's a first look into a new technology for most of us. So we are here because of that, let me say, kind of ugly user interface.

[00:01:07]
So you know that the ChatGPT user interface is not like, wow, but that's not what we see here, we see the power of something totally different behind scenes. So it's not about the why. In fact, this is not a course about ChatGPT. Actually, we won't be using ChatGPT.

[00:01:31]
So ChatGPT, so chat.openai.com is not actually what we're going to cover today. We will see how can we use that as web developers? So I had that question a couple of months ago, as maybe some of you are having right now. Okay, I'm a mobile developer, I'm a web developer.

[00:01:51]
I know that we have GPT. I mean, what's the point? Should I do something about that? Do I need to learn machine learning to actually use some of this stuff? Well, that's why I have created this course to give you a quick look into ChatGPT from a perspective of a web developer.

[00:02:14]
First, ChatGPT, what's that, in case you are here and you haven't heard about it? So it's a large language model designed to generate human-like responses to natural language input, using the GPT architecture. So we have here natural language input. So when we talk to ChatGPT, we need to use natural language, our human natural language, so it's not a programming language.

[00:02:43]
Another definition, ChatGPT is a computer program that can understand and respond to written text like a human would. So it's trying to impersonate us, actually. And the final definition as if we were child, that ChatGPT is like a talking robot that can talk with us by reading what we write and answering us in a way that sounds like a person talking.

[00:03:09]
And actually, all these definitions are coming from ChatGPT itself. And that's the goal of ChatGPT, talk, so it's a talking parrot, okay? I'm not going to get today at least into the details on how it works. We want to use it, we want to understand ChatGPT and use it.

[00:03:34]
So this is not an artificial intelligence course in terms of we're not going to learn how to create a model like ChatGPT. We're going to use it, so we need to understand the concepts and use it in our favor. That's the idea, to use GPT, LLMs, and I will explain what that is in a minute, in our own web apps.

[00:03:57]
Today, we're going to do some cold practices in JavaScript. So the only thing you need to know to follow this course is JavaScript and HTML. And on the server, we are also going to use JavaScript in a very simple Node.js Express Service that we will provide to our frontend.

[00:04:19]
I already have the frontend, so you don't need to create HTML and CSS. Today, we will focus specifically on ChatGPT. So I'm also not here to tell you that ChatGPT will replace all of us, okay? That's a different discussion for a different day. So we are going to use ChatGPT in our apps.

[00:04:41]
So you will see how exciting it is to see all the opportunities that you have to enhance your web apps using the GPT APIs. So we will talk about that. We will talk about how as a web developer, you can today start integrating ChatGPT in different ways. So we will talk about ChatGPT plugins, in case you're interested in that.

[00:05:10]
We will talk about what happens with ChatGPT browsing the web, and not just ChatGPT, also Bing Chat. We will discuss the difference in a minute. They can browse the web, so they can browse our websites. So should we do something about that? Can we optimize the content for AI?

[00:05:28]
Should we do that? Well, we will discuss that as well. But let's say that 80% of the workshop, 80% of this course, it's actually about coding apps that will talk to the, for now, let's say ChatGPT API, then we will become more specific with the terms. Okay, so we will cover OpenAI and ChatGPT.

[00:05:54]
We will try to understand the whole ecosystem that we have today. And you probably know this, but AI is changing really, really, really, but really fast. So everything that I'm saying today may be a little bit outdated in a couple of months, I'm not even telling you years.

[00:06:13]
So that's why you need to rush here and actually start working with this, because everything is changing. So the ecosystem will probably change after a while. Then we'll get into concepts that won't change so fast. I'm talking about how we can use these tools with web development. We will talk a little bit about a concept known as prompt engineering that I will give you a spoiler.

[00:06:40]
I don't like the term, and I will explain why later. But anyway, it's a term that is there, and it's something important that we need to understand. And at the end, we will talk about embeddings and fine tuning. So we will see what is that in case you want to, let's say, for now, customize ChatGPT with your own data, that's the idea.