دانلود کتاب راهنمای توسعه‌دهندگان برای ادغام هوش مصنوعی مولد در برنامه‌ها

Programming isn’t going away. Even the most sophisticated AI system doesn’t spring from the womb with the ability to interface with networks, user interfaces, and databases. Software must be written to bridge the gaps between humans and AI, and between AI and computing platforms (or operating systems). That’s what this book is about. The code that bridges the gap between machine and AI is written in the programming languages we all know and love. Web UIs, backend servers, shell scripts, mobile apps, and drivers all play a part. The code that bridges the gap between AI and humans is, increasingly, the “prompt.” Prompts aren’t just casual chitchat, as you’ll see repeatedly in this book. Prompts are intentionally worded and carefully crafted instructions aimed at accomplishing specific tasks. In other words, prompts are computer code. Much of the job of integrating generative AI into traditional algorithmic software involves writing prompts for a large language model to respond to. The choice of model, how you craft the prompt, and the settings of several parameters influence the quality of responses you get out of a model. Unlike traditional software, however, generative AI isn’t deterministic. Just like working with another human, there’s no surefire way to always get the best, or even the same, results when you give generative AI a prompt. This is where AI engineering comes in. AI engineering focuses on developing, deploying, and maintaining AI systems by applying engineering principles to AI technologies. In this book, we’re looking mostly at the “developing” part of the role. How do you write the code to bridge the gap between humans and AI?