On Twitter/X I post focused content on software engineering and leadership. Where did I get it right? Where did I get it wrong? Let me know.

Follow me on Twitter/X

Prompt Engineering for Humans

Unless you’ve been completely disconnected from recent tech advancements, you’re likely aware of the surge in generative artificial intelligence systems (GAIS) like ChatGPT, Stable Diffusion, Bard, Claude, and Midjourney. These systems, at their core, comprehend text and generate new content in forms of text, computer code, images, or videos. The current state-of-the-art in GAI leverages humanity’s collective works, introducing a new era of content analysis and generation. The quality of results is determined by the clarity and quality of the questions we ask and demands we place on the system. Asking poorly worded questions, omitting details or jumbling too many things into a single request will produce low quality results. On the contrary, learning to adapt requests so the system can shine will produce stunning results that were once thought only possible in science fiction. This leads to an intriguing question: How can we apply the strategies for maximizing outputs from these AI systems to enhance our human interactions and experiences?

Prompt engineering has emerged as a critical skill in this AI-dominated landscape. It involves fine-tuning interactions with GAIS to achieve accurate, precise and efficient results. A prime example is the Midjourney Showcase page, where users continuously refine their prompts to transform the mundane to a masterpiece. This disparity in outcomes between a novice and an expert is reminiscent of the difference in creations using tools like Photoshop and Illustrator, where the same tools can yield vastly different results based on the user’s expertise.

powered by TinyLetter

The burgeoning field of prompt engineering is as fascinating as it is perplexing. In the tech world, there’s a tendency to prioritize machines over human interaction. The “computer nerds” – and I include myself in this group – have enthusiastically embraced learning to work with GAIS. But what if the key skills for successful prompt engineering are the same ones that enable us to excel in teamwork and relationships?

Effectively engaging with both AI systems like ChatGPT and DALL-E, and in our personal and professional relationships, requires thoughtful interaction. This isn’t a far-fetched idea, considering GAIS are trained on human creativity. Hence, connecting with a GAIS can be quite akin to connecting with another human being.

Just as a prompt engineer meticulously considers word choice and sentence structure to elicit the desired response from an AI, similar attention to detail is crucial in human interactions. Whether you’re a friend, leader, coach, or just a likable team member, your success often hinges on your ability to connect with others. Dale Carnegie’s classic “How To Win Friends And Influence People” could very well be rebranded as “Prompt Engineering Humans for Dummies.”

This message is particularly pertinent for tech professionals who often view human elements as secondary. If you find excitement in crafting a new prompt for Bard, consider channeling similar enthusiasm into forming new human connections. Beyond coding and technical skills, fostering strong relationships is vital. The typical software engineer might be adept at prompt engineering for AI, staying abreast of the latest in database technology and cloud APIs, yet they often invest minimal effort in connecting with their peers and managers. Borrowing a phrase from sports, “it’s about the Jim’s and Joe’s, not just the 1’s and 0’s.”

Prompt engineering with humans can manifest in various ways. How we communicate with others significantly impacts the responses we get. Phrasing questions thoughtfully can lead to better and more straightforward answers. Whether it’s motivating a team during crunch time, delivering a halftime speech, or simply interacting with family members, the effectiveness of our communication is paramount. Children, for instance, know exactly how to “prompt” their parents to give in to their demands The way we approach conversations can either build bridges or create barriers. Getting poor results from ChatGPT is a signal that the prompt needs refined or discard altogether. It’s not rocket science; the user is accustomed to finding a new prompt to forge ahead. Your colleague, family member, neighbor and angry customer are all similar in that with the right prompt a connection can be created.

Is authoring a Manager README the correct decision for you?

My next post: Creating The Manager README

I urge you to recognize the parallels between prompt engineering and effective human communication. Acknowledging this connection invites us to reconsider our approach to conversations, both with AI and with each other. Just as a well-crafted prompt can yield remarkable results from an AI, the right words and approach can enhance our human connections, foster understanding, and lead to more meaningful interactions.

The analogy between AI interaction and human relationships transcends mere technical curiosity; it has real-life implications. As technology increasingly mediates our world, the skills of prompt engineering metaphorically translate to the art of conversation and connection. This perspective encourages us to reflect on our communication styles, how we listen, and how we respond, both in digital and real-life interactions. Whether you’re a tech enthusiast, a professional in the field, or simply navigating modern communication complexities, this insight serves as a powerful reminder of the impact of our words and the potential of our interactions. Let’s apply the lessons from the realm of generative AI to our most valuable asset: our relationships with others.

Published on November 25, 2023