The Prompt Safari: Journey Through the Art of Elite Prompting

What happens when a prompt isn’t just a request—but a map, a compass, and a vision?

Our ability to express intent—clearly, creatively, and semantically—has become the new high-performance language of leadership in AI prompt craft. Recently, I went on what I call a Prompt Safari: a multi-step, iterative journey through high-stakes prompting powered by curiosity, strategy, and the poetic precision of words.

The journey and discovery patterns I share here are real. However, in respect of client confidentiality, the use case has been veiled. While the methods, prompts, and models are authentic, the audience and application have been adapted. This article is written not to reveal the work—but to share the structure, mindset, and lessons learned so that others may benefit.

Let me tell you the story—and show you how you can do the same.

This article was originally published on LinkedIn by Severin Sorensen and has been approved for placement on Arete Coach. Scroll to continue reading or click here to read the original article.

What Gets Measured Improves: A Prompt Review Framework

Prompting isn’t magic. It’s a skill. And like any craft, the more we assess and refine, the sharper it becomes. At the end of each day, I ask AI for a rating of how I did with my prompts, requesting feedback on what I did well, what could be better, and how I can improve to reach excellence every time. This self-review loop reinforces clarity, deepens my awareness, and sharpens my craft one interaction at a time.

To help myself and others grow, working with AI, I developed a simple scoring framework that anyone can apply:

Prompt Review Criteria (5.0 Scale)

Rate each prompt across the following five dimensions using a 5.0 scale. Use this guide to evaluate and improve your prompt craft—or to teach others how to elevate theirs.

Five core dimensions to evaluate and improve your prompt craft. Use this guide as a self-assessment tool—or to coach others toward precision, clarity, and strategic AI interaction.

Setting Out on the Journey: From Intent to Precision

My original goal was to build a dynamic contact-sourcing system—for elite construction and project managers who oversee complex residential and commercial builds. These were high-value candidates who might be a fit for leadership roles or strategic introductions across multiple markets.

But in the world of AI, what starts as a search becomes a symphony.

I began by defining the parameters—project scope, location, sector (residential/commercial), experience level, certifications, and even management style. But then came the prompts that refined the strategy:

"Can your identify distinquishable markers of excellence in candidate backgrounds?"
"Can you rank candidate readiness to be contacted or recruited?"
"Are there semantic tells—like 'generational project experience' or 'boutique builder'—that show future potential?"
"How can we score readiness to leave or openness to collaboration based on firm language, size, or affiliations?"

From this, we built a new score: the Green Shoot Readiness Score (GSR) — a signal model for candidate prospecting readiness.

Teaching the Machine, Training the Mind

We didn't stop at finding names. We designed a custom GPT template that accepts a location (e.g., city or zip code) and returns 50 curated candidate profiles with signals like:

Construction specialization
Career journey progression
Management history and longevity
Collaboration openness
Presence in industry groups or elite firms

Then it asked: Are we on the right track? — and adjusted based on user feedback before scaling up to complete our candidate pipeline.

Prompting became architecture. Each instruction became part of a larger system. The results were magical—worksheets of high-potential candidates from any geography were generated within moments. The paradigm of traditional sourcing was broken. Candidate access was now available at the asking.

The task was done, but I was not done. I paused and flipped the mirror: I asked the AI to score my own prompts using the review framework above. What emerged was a feedback loop of excellence—and that loop is available to anyone willing to engage with intent.

Examples to Learn From: Prompts Reviewed & Improved

Let’s apply the framework. Below are actual prompt styles from this safari, generalized for learning and adapted to the talent-sourcing use case. Use these 5 dimensions to review your prompts or teach others to improve theirs.

There were many more prompts in this journey, each one offering new insights, iterations, and learning moments. But you get the picture: Prompt. Iterate. Refine. Reflect. Improve.

Use these examples not as a script, but as inspiration—models to shape and sharpen your own prompting journey.

Why I Say “Explore With Wonderment”

People sometimes ask why I use phrases like:

Explore with wonderment — Opens creativity in AI collaboration
With deep curiosity — Invites thoughtful, layered exploration
Enter with a beginner’s mind — Invites a layered stepwise learning journey approach
Framework + Format + Function — Core structure in high-value prompts

Here's the answer: These words frame the conversation. They invite AI to operate in a more creative, collaborative space. They set the stage for curiosity—and in turn, better results.

Prompting isn’t about commanding a machine. It’s about designing the conditions for insight to emerge.

From My Journey to Yours

The truth is, this wasn’t a story about me being a great prompter. It’s a story about how you can become one. Use these frameworks. Refine your intent. Let curiosity lead the way.

Start with:

What am I trying to learn, uncover, or build?
How can I express that clearly, with a structured format?
What would a strategist—not a technician—ask?
What would excellence look like in this profession?
What have I not asked, that a subject matter expert might ask?
How could AI take this prompt directive and improve it?

Because the best prompt engineers are not code experts. They’re semantic architects of thought.

Closing Thought

If you’re using AI in your business, don’t just give it a task—give it a journey.

Craft prompts with purpose. Layer your intent. Ask better questions. Build feedback loops. Reflect. Iterate. Refine.

You’re not just automating. You’re designing insight. You’re architecting understanding.

And the real magic? It’s not in the model. It’s in the moment when you bring structure, strategy, and language together with clarity.

Explore with wonderment. Discover with intent. Words shape thought. Structure shapes behavior. Wonderment unlocks wisdom.

And why does this work? Because you're not just whispering to AI — you're orchestrating its impact.