The Systems Engineer’s Guide to Modern Design Thinking

How to design with empathy, clarity, and intelligent use of AI — without losing the humanity that makes great products great.

I got to reread some of my favorite design books to write this post, but honestly? It sucked when I realized how little any of them reflected the current reality of the industry. The ideas were beautiful and clean — Norman, IDEO, Brown — but they didn’t match the messy, political, high-stakes AI-influenced world where design decisions now live.

Inspired by the giants of design before me, I set out this week to create a design philosophy of my own. How hard could it be?

Harder than expected.

Because as I reflected on the projects I’ve worked on — from million-dollar engineering decisions to cross-functional system rollouts — the same pattern emerged: I always ended up being the person nudging teams to be more empathetic, talk to the customer early, generate more than one solution, and validate our assumptions before building anything.

My college capstone is the perfect example of what happens when you don’t.

We produced charts, diagrams, models, and documentation that any engineer would applaud. And when we presented the “final” solution, the customer shrugged.

Did we deliver what they asked for? Yeah. Kind of.

Did they ever implement it? Definitely not.

Because while our solution was technically fine, it didn’t solve their core problem — and it didn’t delight them. If I knew then what I know now, we would have followed the process I’m laying out in this post. We would’ve talked with them early. Checked our assumptions. Built options instead of picking one prematurely. Iterated ruthlessly.

I got a B on that capstone. But in my heart, it was a failure — one that still shapes how I design today.

This article is the design philosophy I wish I had back then — and the one I use now.

Why Traditional Design Frameworks Fall Apart in Real Life

Classic texts like The Design of Everyday Things and Change by Design are foundational. They gave us:

• Human-centered design

• Observation before ideation

• Divergence before convergence

• The “fail fast, learn fast” cycle

• The Double Diamond

• Design as intentional change, not magic

But they assume an ideal world — one where you have time, access to users, psychological safety, and team alignment.

Real life?Not so generous.

That’s why I built this hybrid design framework — drawing from design classics, engineering research, and the newest work on AI-augmented design.

The Hybrid Framework: Human Insight + Engineering Discipline + AI Acceleration

This is a modern design process built for real constraints — one that honors both emotional design and systems engineering rigor.

Below is the full walk-through, with modern research integrated where it actually matters.

1. Start With the Real Problem (Not the One You Were Handed)

Donald Norman said it best:

“Good designers never start by trying to solve the problem they are given; they start by trying to understand what the real issues are.”

Industry reality backs this up. A ScienceDirect study comparing expert designers to beginners found experts spend significantly more time in problem scoping before modeling — gathering more information, exploring adjacent issues, and considering more contextual objects in the system.

Translation:If you jump to solutions early, you’re already losing.

Your checklist:

• Talk to the user. Then talk to the user again.

• Ask what they’re trying to accomplish, not what feature they want.

• Look for the political constraints they won’t say out loud.

• Map out what “success” emotionally feels like for them.

This is where emotional design actually begins — not in color palettes, but in making people feel understood.

2. Expand Generously Before You Converge

IDEO’s David Kelley famously pushes designers to “fail fast” — but the important part people forget is that failure is only meaningful after exploration.

Norman describes this as generating ideas “without regard for constraints,” even questioning the problem itself.

Your checklist:

• Create 5+ possible approaches — even if only 2 seem viable.

• Force yourself to produce options your boss did not ask for.

• Separate feasibility from creativity in this step.

This is also where AI shines.

How AI accelerates this step

Modern research (like Pinnacle Pubs’ 2023 “Application of AI Technology in Improving Design Efficiency”) shows:

• NLP models can extract real user needs from massive feedback sets.

• Generative design algorithms can independently create dozens of design variants.

• AI expands the boundaries of what designers think is possible.

But AI does not replace creativity. It multiplies the space you can explore.

Human intuition still leads.

3. Converge on Needs (Not Solutions)

This is where the Double Diamond narrows — but not into a final solution.

It narrows into validated need statements.

Validate the need with:

• Short interviews

• Mock-ups

• Contextual inquiries

• Guerilla testing

• Internal reviews with stakeholders who weren’t in the original room

As Norman notes:

“There is no substitute for direct observation of and interaction with the people who will be using the product.”

Your job is to ensure stakeholders agree on the problem.Once they do, the rest of the project becomes a lot less painful.

4. Expand Solution Options (With Human + AI Co-Creation)

Now that you understand the real need, it’s time to open the funnel again.

What this looks like:

• Multiple prototypes

• AI-generated alternatives

• Storyboards

• Systems diagrams

• Feasibility explorations

AI becomes a design partner, not the designer.

Generative design algorithms help you:

• Rapidly produce variations

• Simulate constraints

• Predict implementation barriers

• Compare options objectively

But research is clear: The core creativity still comes from humans.

AI supports. You lead.

5. Converge on the Right Solution (Using Expert Behavior Patterns)

Engineering design experts — according to the ScienceDirect timeline study — do something unique:

They transition frequently between steps, refining and trimming options quickly. They touch on many ideas briefly, then spend significant time on the few that matter.

Your process should look like a stream: moving forward, doubling back, reevaluating, flowing toward clarity.

Bring in Jakob Nielsen’s heuristics here:

As you converge, evaluate your solution against:

• Error prevention

• Visibility

• Consistency

• Flexibility

• Recognition over recall

• Aesthetic minimalism

• Clear recovery paths

• Clear documentation

This is the “craft” part of design. This is where solutions become usable.

6. Test, Learn, Iterate — Fast

You don’t need a 40-person testing cohort.

Norman recommends testing with groups of five at a time, in both the problem-finding and problem-solving phases.

Your checklist:

• Test early. Test scrappy.

• Let users break your prototype.

• Treat failures as data, not verdicts.

• Iterate publicly so stakeholders feel included.

Iteration builds trust — and trust is the secret engine of every successful design community.

7. Communicate Relentlessly

Here’s the part that would’ve saved my college capstone.

From the ScienceDirect study: Expert designers communicate throughout the process — not just at milestones.

Communicate:

• the problem,

• the need,

• the options,

• the tradeoffs,

• the decision rationale,

• the path forward.

Design dies in silence. Design thrives in clarity.

Putting It All Together: A Modern, Realistic Design Process

1. Understand the Real Problem (Empathy + Context)

2. Explore Options (Human Creativity + AI Expansion)

3. Converge on Validated Needs (User Alignment)

4. Generate Solutions (Prototyping + AI Acceleration)

5. Evaluate & Trim (Expert Reasoning + Heuristics)

6. Test & Iterate (Fast Cycles + Learning)

7. Communicate Constantly (Stakeholder Alignment)

This hybrid model respects the timeless wisdom of Norman and IDEO and the practical realities of engineering and AI-augmented design today.

Why This Matters to the Design Community

We’re entering a world where designers are expected to work faster, produce more, and collaborate across disciplines we were never trained for. AI is becoming a legitimate design partner — but emotional intelligence, problem framing, and human judgment still define great work.

If you’re in engineering, UX, product, or systems design, grounding yourself in a modernized, human-centered, AI-literate process is what keeps your work relevant.

If you want more articles on emotional design, human-centered engineering, and AI-augmented creativity, join my design community.

Subscribe to the blog, share your thoughts, and let’s build a more intentional design world — together.