I forgot to pack for it once, so I drove to the local Walmart and bought what I needed to stand inside Walmart's own freezer: three layers, two pairs of socks. It doesn't matter. A few minutes in, the cold stops being weather. You feel it in your bones.

I learned to use that. When suppliers came to visit, I liked to walk them into the freezer and talk price right there. Everything's a negotiation, and nobody stretches out a meeting at fifteen below.

People worked full shifts in that room. The job was called order filling: drive a forklift through the racking that holds our food and ice cream, stop slot after slot, hand-build the pallet a store ordered. Bending and lifting for hours in air cold enough to freeze your eyebrows. One of the hardest jobs in the company.

I spent nine years as an automation engineer at Walmart, and I helped design the system that replaced that job with robots. None of it is secret. Search "Walmart Witron" on YouTube and you can watch it run.

Nine years of that leaves you with convictions. These are the hills I'll die on, from someone who designed multi-million dollar automation and would rather help artists.

Some jobs should never have existed for humans

Our bodies weren't designed to be industrial athletes. People got used to the freezer, sure. Some genuinely liked it, the pay was good, and it's still work. Right now, while you read this, somebody is building a pallet in that cold so the shelves stay stocked. Kind of beautiful and dystopian at the same time. I won't pretend companies automate that job out of kindness, and I won't pretend anyone should have had to do it either. Both are true, and I stood in the middle of it for nine years.

Automate the invoice. Never the art.

Just because you can automate something doesn't mean you should. You could automate selling artwork online. The happier route is automating the boring parts of the business and spending the hours it gives back making things. Right now my hours go to 3D printed gear for music fans. Next year, maybe something else. That's the point.

You're not scared of technology. You're scared of your system failing you.

People will be displaced, and people will adapt. We always have. The transition is the part we have to get right: learning new skills, and pushing our systems to change with us instead of leaving people behind. Traditional information work will shrink, and I think that can be a good thing. Humans should be creating.

Automation is creating a new type of artist

A thinker with ideas to share who never clicked with the old mediums. Maybe their thing is 3D printing. Maybe it's curated media that tells a story instead of hacking your attention. Blindly generating slop until something sticks is wasteful. Curating tasteful, intentional work is still art. Slop is still slop.

If you won't stand behind it, that speaks volumes about your intention

Some faceless work is great. Sometimes the idea hits harder without a face, and I love those. The tell is when someone is selling you something and won't put anyone real behind it. AI influencer culture, fake beauty and hacked psychology with nobody home. That's the intention showing.

I'm just an engineer learning in public. Some of this is backed by data, some of it is opinion, and I think that's OK. Let's reverse engineer a better life.

The system was a 30,000-unit-per-hour apparel sortation line. The facility said they were capacity constrained and asked for an expansion. That request meant real capital, real lead time, real risk, and it would have gotten approved on the strength of the ask alone. Nobody questions a facility that says it's out of room.

Before signing off, we mapped every loss on the existing system instead of the request in front of us. What we found: staffing gaps and downtime were keeping the system from anywhere near its real ceiling. When it was properly staffed during peak periods, it could already handle far more volume than anyone believed. There was no urgency to run it well, because everyone had already decided the fix was more hardware.

The actual fix was an operational playbook and a seasonal staffing strategy. No new automation. It avoided close to $1M in unnecessary capital investment.

The lesson, and the thing I bring to every organization I work with since: the expensive answer is usually the easy one to say yes to. The right answer is almost always sitting inside the system you already have, if someone's willing to go find it before reaching for the checkbook.