title: "An anthropology degree was the most useful thing I brought to engineering" slug: 'anthropology-degree-engineering' date: '2026-04-15' excerpt: "The major engineers laugh at quietly turned out to be the edge." tags: ['career', 'identity', 'humanities'] category: 'Career' published: true
When I tell people I have an anthropology degree, the response is usually a small wince followed by some version of "huh, that's interesting, how'd you end up in tech." Polite. Slightly puzzled. Sometimes condescending if the person is younger than me.
I used to be defensive about it. Now I think it was the most directly useful thing I did in college.
Here's the unflattering truth: nobody learns to be a senior engineer in school anyway. The CS majors I work with had to learn the actual job after they got hired, same as me. What school gives you is a set of habits of thought that you'll apply to whatever you do next.
The habits anthropology gave me:
Take fieldwork seriously. When you walk into a new system — a codebase, a team, an enterprise client's organization — you are doing fieldwork. The thing that is true is what people actually do, not what they claim they do, and not what the documentation says. Senior engineers who skip this step get expensive surprises. Anthropologists are trained to default to fieldwork.
Watch for ritual that nobody can explain. Every team has rituals. Some of them have reasons that are still load-bearing. Some of them are residue from a constraint that disappeared three years ago. The trick isn't to throw all of them out; the trick is to ask gently and see which ones get a real answer. The ones that don't are usually safe to retire. (FedNow had a fourteen-day review step that turned out to be fossilized residue from an audit requirement that had been resolved years earlier. Killing it saved us a third of the schedule.)
Resist your own model. Engineers are trained to abstract. We see a problem and immediately reach for a pattern we already understand. Anthropologists are trained, painfully, to do the opposite — to hold off on naming what you're seeing until the data has had a chance to surprise you. About half the production bugs I've shipped were because I assumed I knew the shape of the problem before checking.
Note the gap between what's said and what's done. This one is gold for working with stakeholders. People will tell you what they want. They will then behave in ways that contradict what they told you. The contradiction is the actual signal. If you can see it without judgment, you're already in the top quartile of cross-functional collaborators.
I'm not saying every engineer needs an anthropology degree. I am saying that the skills I picked up in those classes — observation, ethnographic patience, suspicion of one's own first-pass interpretation — have made me money, kept me out of trouble, and gotten me promoted faster than my CS-degreed peers in roughly equal proportion to the technical skills I learned later.
Engineering schools could stand to teach more of this. They mostly don't, because nobody's writing leetcode problems for "is this stakeholder telling you the real constraint." The good engineers learn it on the job and the great ones bring it in from somewhere else.
I happened to bring it in from anthropology. Other people bring it from theater, from the military, from years as a parent. The substance is the same. It's the part of the job that isn't code.