I spent five years in fine dining. Including two years spent as a pastry chef, which means I spent well over ten thousand hours obsessing over grams, temperatures, the ways to slice a steak, and how best to fill a cream puff. I have designed dozens of menus and plated hundreds of thousands of dishes.

A few years ago I made a transition into IT Consulting, so today, my tools look different. I now support Google in developing focused AI use cases and spend my time in the world of Business Intelligence and AI based Automation.

On the surface, these two worlds have nothing in common. One is loud, hot, and smells like toasted sugar. The other is quiet, remote, and smells like expensive coffee. But the thread that connects them is the value of a key skill: discipline. Which is defined by systematic thinking and a relentless focus on creating what the customer truly wants.

Most of the “AI experts” you see online right now are trying to sell you a miracle. They promise that AI will change everything overnight, replace your entire staff, and print money while you sleep. They are selling hype.

I’m here to tell you that AI is just a kitchen tool. And like any tool, it’s only as good as the cook using it.

Slow is Smooth, Smooth is Fast

There’s a phrase you hear in professional kitchens: “Slow is smooth, and smooth is fast.” The saying has roots in military training, and the kitchen borrowed it honestly; the modern brigade system itself came from the military. But nowhere does the principle prove itself more faithfully than in a working kitchen.

Here’s what it means: when you rush, you make mistakes. You drop things, you burn things, you have to start over. But when you move deliberately, when you focus on doing each step precisely, you develop an infallible rhythm. Your movements become efficient. Waste disappears. And suddenly you’re faster than the person who was rushing, because you’re not spending half your time fixing mistakes.

The phrase isn’t about moving slowly. It’s about moving correctly. Speed is a byproduct of technique, not a replacement for it.

Early in my kitchen career, I had to learn to separate eggs. Not two or three for a home recipe, but dozens at a time for pastry production. At first, I just focused on learning the technique. Crack the shell cleanly. Use my hands to let the whites slip through my fingers while keeping the yolks intact. Don’t rush. Don’t break yolks. Don’t get shell fragments in the bowl, because fishing them out costs more time than you saved by going fast.

After a few weeks, the head chef started timing me as a joke. The thing was, I had actually gotten fast. Not because I was trying to beat a clock, but because I had stopped making mistakes. No broken yolks to discard. No shells to fish out. No wasted product. The speed came from the smoothness, and the smoothness came from learning to do it right in the first place.

This is the same reality we are facing with AI. Companies are rushing to implement tools they don’t understand, automating processes that aren’t ready, and then spending months cleaning up the mess. The ones who will win are the ones who slow down long enough to build correctly and then move faster than everyone else because they’re not constantly fixing their mistakes.

Why Your AI Strategy is Probably a Bad Menu

Most people believe the majority of restaurants fail. You’ve heard the statistics. But actually, UC Berkeley data suggests only 17% fail in their first year. Those that do close don’t fail because people don’t want to eat. They fail because the menu is something people don’t want, or the price is too high, or the service is pretentious and annoying. Most restaurants succeed because they fill the niche that guests want.

The same is not true for AI. Thousands of AI use cases have failed in the last two years. According to an MIT study, only about 5% of AI implementations are actually delivering real value. They fail because companies build tools no one asked for, or they try to automate things that require human judgment, or they get so caught up in the “tech” that they forget the human consumer at the end of the line.

In a kitchen, the customer buys what they like, not what the chef wants to make. You have to listen to feedback from people who aren’t chefs. In AI development, your users don’t care how the model was trained or how many parameters it has. They care about what it delivers. They care if it solves their problem and how often they can afford to use it.

A flashy dish that no one orders is just wasted prep time. A flashy AI tool that no one uses is just wasted development budget. Both problems have the same root cause: building for your ego instead of for your customer.

The Sous Vide Theory of Augmentation

People are terrified that AI will cause mass unemployment. I think that fear is exaggerated. We aren’t close to “superintelligence.” We are decades away from machines that can truly rival the breadth of human intelligence, mostly because we have massive physical computing, energy, and storage hurdles to clear first.

The real value of AI right now is in specialized tools. Think of it like a sous vide machine.

A sous vide allows a cook to bring a steak to the perfect temperature with zero human intervention. It doesn’t replace the chef. It just ensures that the longest, most repetitive part of the process is handled perfectly every time. This reduces the “time to plate” for a steak from twenty minutes down to five.

The human is still there to season it, sear it, and plate it. The human provides the judgment, the emotion, and the final touches. The machine handles the part that was tedious and error-prone. The human handles the part that requires finesse.

Don’t look for 100% task automation. That’s a trap that leads to rushed, poor-quality work, the opposite of smooth. Instead, look to reduce your workload by 50-75% and keep a human in the loop to make sure the final product is actually worth serving.

What This Content Will Deliver

I’m writing this because I want to offer a different path. One that isn’t built on “AI evangelism” or corporate jargon. I want to help people with zero technical background see that they already have the skills to navigate this shift. If you know how to follow a recipe, manage a station, or plan a menu, you already understand systems thinking.

Every Friday, I’ll be sharing one post from a rotating menu:

Concept Deep Dives: I’ll take a kitchen metaphor, like mise en place, and show you how it applies to things like prompt preparation and data hygiene.

Trend Analysis: I’ll look at the latest AI news through a kitchen lens. No hype, just a look at what’s actually useful and what’s an inedible garnish.

Case Studies: Real stories from my consulting work where we’ve actually made things smoother and faster.

Dish/Model Applications: Specific tools and workflows you can actually use today.

Why You Should Follow Along

If you are intimidated by AI, you should care because you already understand these concepts. You just need someone to translate the “tech-speak” into something that makes sense.

If you are already in the tech industry, you should care because kitchen wisdom will make your work smoother. It will help you ship better products that people actually want to use.

If you are a business leader, you should care so you can learn to spot AI “snake oil” the same way you can spot a bad dish on a menu.

AI is changing millions of lives. For that change to be positive, we need to stop looking for magic and start looking for discipline. We need to learn the technique before we chase the speed. We need to do it slow, do it right, and only then do it faster.

Because slow is smooth. And smooth is fast.

I’ll see you on the line next week.

Jacob