Stop Hiring Prompt Engineers, instead look for Systems Thinkers

“The AI skills gap isn’t technical. It’s systemic.” — paraphrased from @alliekmiller

Let’s talk about the elephant in every AI transformation room:

Companies are chasing LLM experts like it’s 2022. They’re asking: “Can this person code? Can they prompt ChatGPT?” But the real question should be:

Can this person think in systems?

If you’ve ever tried to plug AI into a broken process, you know how useless it is. AI doesn’t magically fix bad workflows. It amplifies them. It scales dysfunction unless paired with systems thinkers—people who understand business processes, AI potential, and the mindset to grow.

Let’s break this down.👇

Who is a Systems Thinker?

Systems thinking has been discussed for a long long time! I have taken systems thinking classes more than a decade ago from Prof. Sanjiv Vaidya and Prof. Priya Seetharaman . I loved the simplicity back then and I feel it is super applicable now in the age of AI. Systems Thinker sits at the intersection of:

1. AI Knowhow – Not just prompt engineering, but understanding LLM capabilities, limitations, and best practices.

2. Business Context – How a process works, where the friction is, what the levers of ROI look like.

3. Growth Hacking Mindset – A bias toward learning, iterating, and scaling.

When these three intersect, you don’t get a techie or a traditional manager—you get someone who knows how to deploy AI in the real world.

🛠️ Why Prompt Engineers Alone Won’t Cut It

In the past few months, I’ve met brilliant prompt engineers who can make GPT-4 dance. But that’s not enough. In fact just today, I met an engineer who was sharing a similar frustration - 'Writing prompts isn't the end, it's the beginning!' How prophetic indeed! In fact, here's a shoutout to her for inspiring this post! (Name withheld on request) . Why is it not enough -

Here’s why:

• A well-prompted LLM can summarize a PDF. But if you don’t know why that summary matters to your sales funnel, it’s just text.

• An AI image tool can generate 100 creatives. But unless you know which funnel stage needs what kind of message, it’s just digital clutter.

• You can generate 500 cold emails with AI. But unless your CRM workflow supports lead scoring and follow-up—you’re burning leads, not building them.

AI needs context. And context comes from systems thinkers.

Real-World Example: Turning Chaos into Campaigns

Last quarter, we tested an AI-powered content generation stack: Midjourney + ChatGPT + Claude 3.7. We aimed to reduce the creative production time for Instagram carousel ads.

The technical team delivered. The AI team generated content. But performance was flat. Why?

Because the campaign didn’t integrate into the overall product funnel -

1. No hook to onboarding.
2. No incentive to install.
3. No CTA that mapped to the growth goals.

Once we got a systems thinker involved, the whole engine came alive:

• We mapped content to intent stages (TOFU, MOFU, BOFU).

• Used Claude to analyze sentiment from past campaigns.

• Automated feedback loops into Meta Ads Manager.

Result? Early trends do look promising! Lets see how it sticks!

🎓 Who Can Become a Systems Thinker?

Great news: Systems thinking isn’t a job title—it’s a skill set.

And like any skill, it can be taught.

Here’s a starter toolkit for developing system thinkers inside your org:

The Systems Thinker Playbook - a simple three step filtration hack

Let’s say you’re hiring someone to lead your AI content automation. Here’s what to ask:

1. How do you integrate AI into a broken workflow?

— Look for answers around diagnosis, not just tools.

2. Can you define what success looks like before generating content?

— This reveals strategic alignment.

3. How do you update your system when the model improves?

— AI is a moving target. Systems thinkers plan for evolution.

Curiosity as a Prime Mover for Systems Thinkers -

Who are the architects, product leaders, visionaries? Everyone faces hurdles in their career. The ones who thrived weren’t the best coders or the best sales people. They were the ones who asked:

• “Where does this fit in the bigger picture?”

• “How will this impact the end-user?”

• “What’s the next step after we deploy this?”

That curiosity—that urge to see the whole system—is what separates the AI rockstars from the one-prompt wonders.

💼 Final Thoughts / Summary for Founders and Managers:

✅ Don’t just hire AI experts. Hire people who understand your processes.

✅ Train your ops and marketing folks in basic AI tools.

✅ Start cross-functional squads to run small AI experiments.

✅ Reward understanding over output.

#AI #SystemsThinking #FutureOfWork #Leadership #PromptEngineering #BusinessStrategy #InsightEdge #SiddheshJoglekar #Claude3 #ChatGPT #DigitalTransformation #Hiring