Prompt Engineering Could Be the Hottest Programming Language of 2024 — Here’s Why | by Nabil Alouani | Dec, 2023

Large Language Models are the next generation of Operating Systems

Nabil Alouani
Towards Data Science
a woman with dark skin and curly hair wearing glasses, seated at a desk in a dimly lit room, illuminated by the soft glow of her laptop screen. Papers and digital devices are spread out on the table, suggesting a busy work environment. The woman appears focused and engaged with her work, possibly coding or analyzing data, as indicated by the graphical elements on her laptop screen. The setting conveys a late-night work atmosphere with a sense of calm concentration.
Unless indicated otherwise, all the images are generated by the author using Midjourney, DALL-E, and Canva.

“I don’t think it’s accurate to think of Large Language Models as chatbots or some kind of word generators,” Andrej Karpathy, one of OpenAI’s founding members, said. “It’s a lot more correct to think about [them] as the kernel process of an emerging Operating System.”

Wait, but what the heck does that mean?

Large Language Models (LLMs) will gradually become the interface between you and computer systems.

Right now, you’re holding a device that has some computing power inside it, but you can’t directly access that power. Your interaction is mediated by an Operating System (such as Windows, Mac OS, and Android,), which transforms a collection of chips and circuits into a user-friendly interface.

Your Operating System (OS) allows you to perform a wide range of activities (like reading some bald dude’s article) through a variety of apps running on top of it. Each app has its own User Interface (UI) and its own set of tasks it can accomplish. You jump from one app to another, one UI to another, depending on what you need to do.

An infographic illustrating the transition from current operating systems like iOS, macOS, Windows, Linux, and Android to a large language model as an operating system. The top half, labeled “Software 1.0,” shows icons for an Internet browser, image/video generation, Python interpreter, file storage, and calculator, all with user interfaces pointing to a human silhouette. The bottom half, labeled “Software 2.0,” shows the same icons connected directly to a large language model instead of via OS.

Tomorrow, you’ll have a single UI to do everything from writing an annual business report to building a new app from scratch. The said UI will be a chatbox or a “context window” inside which you can submit instructions in natural language — and that’s where Prompt Engineering comes into play.

Prompt Engineering is a fancy way to say “Write better and better instructions for your AI model until it does exactly what you want.” Except, it’s not merely wordplay; it’s the blueprint for the future of programming.

“Programming refers to a technological process for telling a computer which tasks to perform in order to solve problems,” Coursera wrote on their website. “You can think of programming as a collaboration between humans and…

Source link

This post originally appeared on TechToday.

Leave a Reply

Your email address will not be published. Required fields are marked *