To tell the story of how we got here, to AI tools like ChatGPT, powered by large language models (LLMs), let’s first cover natural language processing (NLP).

NLP is a field of computer science, artificial intelligence, and computational linguistics. It’s concerned with the interactions between computers and human language, and how to program computers to process and analyze large amounts of natural language data. NLP is a hugely interesting area that has a range of useful applications in the real world. Here are some:

• Speech recognition: If you have a modern smartphone, you’ve likely interacted with voice assistants like Siri or Alexa, for example.
• Machine translation: Google Translate is perhaps what comes to mind when thinking of machine translation, the ability to translate from one language to another automatically.
• Sentiment analysis: A very useful area is understanding the sentiment in areas like social media, for example. Companies want to know how brands are perceived; e-commerce wants to quickly understand product reviews to boost their business.
• Chatbots and virtual assistants: You’ve likely seen chatbots being integrated on web pages even before the advent of ChatGPT. These chatbots can answer simpler questions, and companies have them to ensure you quickly get an answer to simpler questions and provide a more natural experience than an FAQ page, among other usage areas.
• Text summaries: Search engines come to mind again when thinking about text summaries. You might have seen how, when you use search engines like Bing or Google, it’s able to summarize a page and show the summary together with the link to the page in a search result page. As a user, you get a better understanding of what link to click.
• Content recommendation: This is another important area used by a variety of different domains. E-commerce uses this to present products you’re likely to be interested in, Xbox uses this to recommend what games to play and buy, and video streaming services display content you might want to watch next.

As you can see already, with NLP, both companies and end users benefit greatly from adopting it.

The rise of LLMs

How did we evolve from NLP to LLMs, then? Initially, NLP used rule-based systems and statistical methods underneath. This approach, although working well for some tasks, struggled with human language.

This changed for the better when deep learning, a subset of machine learning, was introduced to NLP, and we got models like RNN, recurrent neural networks, and transformer-based models, capable of learning patterns in data. The result was a considerable improvement in performance. With transformer-based models, we’re starting to lay the foundations of large language models.

LLMs are a type of transformer model. They can generate human-like text and, unlike NLP models, they’re good at a variety of tasks without needing specific training data. How is this possible, you ask? The answer is a combination of improved architecture, a vast increase in computational power, and gigantic datasets.

LLMs rest on the idea that a large enough neural network can learn to do anything, given enough data and compute. This is a paradigm shift in how we program computers. Instead of writing code, we write prompts and let the model do the rest.

GPT models

There are many different types of LLMs out there, but let’s focus on GPT for a second, a type of LLM on which the book’s chosen tools are based (even if GitHub Copilot uses a specific subset known as Codex).

There have been several different versions developed in the last few years. Here are some models developed by the company OpenAI:

• GPT-1: The first one, with 117 million parameters using transformer architecture.
• GPT-2: This model has 1.5 billion parameters and is able to generate coherent and relevant text.
• GPT-3: This model has 175 billion parameters and is considerably better than its predecessor with features like answering questions, fiction generation, and even writing code.
• GPT-4: This model has been quoted to have 1.76 trillion parameters.
• The number of parameters allows the model to understand more nuanced and coherent text. It should also be said that the larger the model, the larger the computational resources that are needed to train it.
• ChatGPT recently switched to GPT-4 and the difference compared to GPT-3 is significant.

How LLMs are better

Now that we have a better understanding of how LLMs came to be and where they came from, what makes LLMs great? What are some good examples of why we really should adopt AI assistants based on LLMs?

Because LLMs are bigger and more advanced, there are some areas in which they clearly outperform traditional NLP models:

• Context: LLMs can understand not just the recent input but can produce responses based on a longer conversation.
• Few-shot learning: To perform a task, LLMs usually just need a few examples to produce a correct response. This should be contrasted with NLP models, which usually use a large amount of task-specific training data to perform properly.
• Performance: LLMs are better than traditional NLP models in areas like translations, questions, and summarization.

It’s worth mentioning that LLMs aren’t perfect; they do generate incorrect responses and can sometimes make up responses, also known as hallucinations. It’s our hope though that by reading this book, you will see the advantages of using LLM-based AI assistants and you will feel the pros clearly outweigh the cons.

Probably the biggest game changer with using LLM-based AI assistants is that you’re able to interact with them using nothing but natural language. There’s no need to learn a programming language to get the response you need. This change constitutes a new paradigm in interacting with AI. We’re moving away from writing in specific languages for producing apps, data retrieval, or even how we produce images, presentations, and more to express at a high level what we want through a prompt.

Here is an example of things that are now possible to do using prompts, where it before needed considerably more effort:

• Programming: With a prompt, you express what app you want to build or what changes you want to make with the code.
• Image generation: Where you before needed a designer or artist, you can now generate via prompts.
• Videos: There are tools out there that, once given a prompt, will generate videos where an avatar reads out your written text.
• Text tasks: LLM-based AI assistants can generate emails, summarize large bodies of text, author interview ads, and much more; anything you can imagine with text really.

All these application areas mentioned above make it clear that LLM-based AI tools are useful not only to programmers and data scientists but numerous different professions.

Challenges and limitations

Is everything working perfectly at this point? AI assistants aren’t able to replace “you” just yet, and should be considered more of a “thinking partner.” Microsoft has even, through conscious naming, called their AI assistants “Copilots” where you’re clearly the pilot that sets out the direction. These tools can generate text and other modalities in seconds, but you need to verify the correctness. Often, the first response you get from a tool is something you need to iterate over. The good news is that it just takes seconds to redo the instruction.

An important thing to realize about AI assistants is that the more skilled you are at a certain topic, the more intelligent questions you can ask of it, and you’ll be able to better assess the correctness of the response.

The goals of this book are to:

• Introduce you to the new paradigm of programming with natural language.
• Provide you with the tools to get started using AI assistants.
• Empower you to use AI assistants effectively and responsibly by teaching you prompt engineering and specifically a set of prompting strategies (covered in Chapter 2) and some sound practices (covered in Chapter 8).

We believe that with these tools, prompting strategies, and practices, you will be able to use AI assistants effectively and responsibly to augment your work and increase your productivity.

This book is for professional developers within both the web and machine learning space. It is for those who want to learn how to use AI assistants like GitHub Copilot and ChatGPT to augment their work and increase their productivity.

Programming has gone through a series of changes and paradigm shifts throughout history:

• Ada Lovelace wrote the first algorithm for a machine, the Analytical Engine, in the 1840s. Lovelace is considered the first computer programmer and the first to recognize that the machine had applications beyond pure calculation.
• In the 1940s, the first programmable computers were created. These computers were programmed using punch cards. One such computer was the Harvard Mark I, which was used to calculate the trajectory of artillery shells. Also, Bombe is worth mentioning, which was used to crack the Enigma code during World War II and was instrumental in the Allies winning the war.
• In the 1950s, the first high-level programming languages were created. This time period saw the birth of FORTRAN, LISP, COBOL, and ALGOL. Some of these languages are still in use today, especially in banking systems, scientific computing, and defense.
• In the 1970s, the first object-oriented programming languages were created. The 1970s meant we got Smalltalk, C++, and Objective-C. Except for Smalltalk, these languages are heavily in use today.
• In the 1990s, the first functional programming languages were created. The 1990s gave us Haskell, OCaml, and Scala. The benefit of these languages is that they encourage immutability and pure functions, which makes them easier to reason about and test.
• In the 2000s, the first declarative programming languages were created. Declarative programming languages are used to describe what you want to do, rather than how you want to do it. The 2000s gave us SQL, HTML, and CSS.
• In the 2010s, the first low-code and no-code platforms were created. These platforms opened programming to a wider audience, and allowed anyone, regardless of technical background, to build applications.
• In the 2020s, the first AI assistants were created that leveraged natural language. If you can write a sentence, you can write code.

In summary, programming has gone through a series of changes and paradigm shifts. Prompt-first programming is the latest paradigm shift and mastering it will be key to staying relevant in the immediate future.

Looking ahead

If changes and paradigm shifts took years or decades in the past, they now take months or even weeks. We’re moving toward a new world at breakneck speed.

There’s reason to be excited, as we’re moving faster than before, but as always, we should exercise caution. We should be aware of the risks and the dangers of using these tools irresponsibly, but most of all we should be aware of the opportunities.

As Alan Kay once said, “The best way to predict the future is to invent it.”

We believe the best way to use this book is to follow the chapters in order.

Chapter 2, with the prompting strategies, is the most important chapter in the book. These patterns and strategies are referred to throughout the book and are the foundation for how to use AI assistants effectively and responsibly.

The book is written in the following format:

• Introduction: The first chapter aims to provide you with an overview of what this book is about, its goals, and who it is for.
• Prompt strategy: The idea is to lay the foundation on how to break down problems within the domains of data science and web development. From this chapter, you will learn strategies you can adopt for your own problems.
• Tools of the trade: The third chapter introduces you to our tools, GitHub Copilot and ChatGPT, what they are, how they work, and how to install them. However, the book is written in such a way that you can take any of the prompts we suggest and feed those into any AI assistant, and get a similar experience.
• The remaining chapters of the book show how we use the prompt strategies from Chapter 2 and apply them to various domains from web development to data science and machine learning.

Happy reading!

Join our community on Discord

Join our community’s Discord space for discussions with the author and other readers:

https://packt.link/aicode