Prompt Engineering from 2020 to 2025 ✨

A Growth hacker turned Prompt Engineering wizard shines a light on this mysterious new role and these essential new applied AI skills. 💡

Hey Everyone,

I’ve received a lot of requests about Prompt Engineering. So I asked one of the world’s foremost experts on it, he even wrote a book about it, see it on Amazon here.

• Prompt Engineering for Generative AI

As you know, Prompt engineering is the process of structuring an instruction that can be interpreted and understood by a generative AI model. Currently there’s a general consensus that Anthropic’s Claude 3.5 is the best at interpreting human prompts.

Mike Taylor

has in recent times also co-written huge guest posts for

Lenny Rachitsky

’s Newsletter. But it’s Mike’s day job that really puts Generative AI to the test each day. Mike co-founded a 50-person marketing agency called Ladder, created marketing & AI courses on LinkedIn, Vexpower, and Udemy taken by over 400,000 people, and published a book with O’Reilly on prompt engineering. He builds AI products at Brightpool.dev.

As I dabble more into applied AI, I’m asking around for those who are more advanced tinkerers. Meanwhile we’re also beginning to explore applied AI tools on this Newsletter by popular demand:

Also Read:

• 🛠️ AI for Non-techies: Meeting AI assistants and Notes.

• 🔧 AI for Non-techies: AI Tools for Email writing and Inbox management.

Since Mike literally wrote the book on Prompt Engineering, I wanted his take on how all this began and where it’s going. This article will break down Prompt Engineering into periods and chart out what’s next. The article becomes available to all readers at a later date.

I actually first discovered Mike’s writing over at Every.io. He writes Also True for Humans, a monthly column at Every, that is specifically an AI and business publication. Check out his acclaimed Udemy course. Mike is a serial entrepreneur and shares his knowledge specific to this topic.

---

By

Mike Taylor

, August, 2024.

🎧Listen to it: 20:01.

Do you know someone curious about Prompt Engineering best practices and its history?

👨 “Also true for humans” 👉

Read more by the Author:

• Why AIs Need to Stop and Think Before They Answer

• Prompt Engineering: From Words to Art and Copy

• How close is AI to replacing product managers?

• Putting a Prompt into Production

• In the article Mike shares his learning through time and his experiences optimizing Prompt Engineering via new models and evolving best practices.

Foundational Prompting Tips

While the field of prompt engineering has evolved rapidly with advanced techniques like chain-of-thought prompting and emotional stimuli, it's important to remember that certain fundamental principles remain effective regardless of the AI model or application. Whether you're a beginner just starting out or an experienced user working with cutting-edge models, these basic techniques will always help you get better results from AI:

1. Be Specific: Be specific in your requests. Rather than asking broad questions, focus on detailing exactly what you need. For instance, instead of asking generally about video conferencing, you might say, "Explain step-by-step how to set up and join a Zoom meeting on a Windows laptop, including how to test audio and video." This level of detail guides the AI to provide more precise and useful information.

2. Break Down Complex Tasks: Break down complex tasks into smaller, manageable parts. When tackling a home renovation project, for example, don't ask about the entire process at once. Instead, inquire about specific aspects like the key steps in planning a kitchen remodel, the pros and cons of different countertop materials, or budget-friendly options for updating a kitchen without full renovation. This approach allows you to gather detailed information at each stage of your project.