The Tech Career You Can't Go To School For

Part I: From Dead-End Cubicle to DevOps

It was the late 2000s and I graduated with a Bachelor’s degree and was working in a nondescript, dead-end office job. I was making a modest living—enough to pay rent but not much else—in a mid-sized city nowhere near a tech hub, with no upward mobility or any obvious way out. The one good thing about this job is that it was easy and I had a computer all to myself for 8 hours a day, so when my paid work was done I could spend hours educating myself and working on side projects.

For the previous 10 years or so, at home I was a Linux nerd. I got started in the early days of the 2.4 kernel and package managers that were glorified tarball unpacks with no dependency tracking. Back in those days there was no GUI email client, so I spent a week getting sendmail MTA set up, and then used fetchmail to download via POP3 into my mutt terminal email client. It was obviously insane and nothing that any normal person would put up with, but I was young, broke and had a lot of time so to me it was fun. I broke my systems constantly just messing around, then spent days fixing them. The odd thing is, I never considered it any kind of career path. In fact, I was kind of embarrassed and was afraid of being labeled a “nerd” because I thought Unix and Linux systems were cool.

That all changed as I sat in my cubicle, in my shirt and tie, bored to tears and terrified that this would be my life until I die. I realized, very belatedly, as I saw how my coworkers and boss interacted with technology, that I had skills that were not typical. Maybe even marketable. But I had a science degree unrelated to computers and no formal tech experience whatsoever. What could I do?

First thing I did was use a fortunate assignment at work to teach myself the beginnings of programming. First using R, then Python. I already had a lot of the sysadmin foundation, I just needed the time and motivation to learn the basics of coding and my job gave me that opportunity. By the end of it, I could put together scripts and even (for me) fairly complex PyQt graphical applications. I wasn’t any good, obviously, but I had the rough beginnings.

I spent the next year—not an exaggeration, a full year—sending out my resume and applying to every vaguely-related tech job posting I could find. The hook in my resume was the 10 years of Linux experience (I was honest about it being unpaid home hobbyist work) and then some perhaps slightly embellished items about my R and Python work. I got a fair number of callbacks surprisingly, but I almost never made it past the recruiter screen. I was obviously totally inexperienced and it showed, but I kept persisting. At times I thought I was crazy and this would never work out, but I kept sending in applications and taking recruiter calls during my lunch breaks.

One day, I got an actual interview. It was with a startup I had never heard of, on the other side of the country, in an industry I knew nothing about, but they were friendly and interested. I made it clear that I was motivated, wanted to work in tech and I was willing to move across the country in 30 days. After a few more calls with them, I got the offer letter for a Junior DevOps Engineer position.

The offer was essentially exactly what I was making in my current job. Except this was in a city with one of the highest costs of living in the country, so it was an effective pay cut. It also was about half of what a typical mid-level DevOps role was going for, in the startup world. But it was a job, in the field I wanted, in an exciting location. I decided to go for it, and packed up everything and drove across the country.

I’m not going to sugar-coat this: it was hard. I’ve never worked as hard in my life as I did for those first few months, trying to prove myself in the pressure cooker of a VC-backed startup of a few dozen people. I was used to a 9-5 office job, not grinding from 9:30AM to 8-9PM every day and dealing with production outages. But that’s what my coworkers did so I followed along. Eventually, I learned some things. Got promoted. Was mentored by the CTO. Things were looking good.

Two years later, now I’m making about 20% more. Definitely better, but even more significantly, now I had actual professional experience and I was living in an area with a real tech ecosystem.

I reached out to some independent tech recruiters, gave them my availability, and to my shock, started getting solid leads. It was surreal to have a recruiter working for me—it felt like having a Hollywood agent or something. She was setting up my interviews and negotiating terms with the companies, it was unreal. I landed a job with a hot, well-known startup and their opening offer was double what I was making, plus equity.

To this day, it’s the largest percentage income jump I’ve ever landed in a single job change, and certainly the most satisfying. This validated the gamble I made a couple years earlier, and turned all my Linux tinkering from a quirky hobby into a solid career investment.

I won’t bore you with what happened next—promotions, startup exits, move into management—but I want you to understand that this career opportunity is still available today for those with the motivation and skill set. Regardless of formal credentials.

Why? Operations skill is provable and demonstrable. You prove you have the skill by doing it. You don’t need a million dollar lab or permission from anyone to practice, you can build skills yourself on your own time using easily-obtainable consumer hardware.

What you need to take away from this: there’s nothing special about me. Although the details differ a bit now some years later, the door I stumbled upon by accident is the same door that’s available to you right now. The story is the same, but you can do it better because I’ve already mapped out how to get there and I’m ready to share it with you and help you step by step.

Let me be real though: it won’t be easy. You have to pay your dues any way you look at it: either the way I did it, over 10 long years of unfocused tinkering + 2 years of intense industry “boot camp” of sorts. Or you can do it faster and better because you’ll know the skills you need and you won’t waste time on things that don’t matter.

In Part II, I’ll share the roadmap as it exists in 2026: focused training areas, roughly how you know when you’re “ready”, where the industry entry points are, and the roles and jobs that can serve as your bootcamp, and the ones that are a waste of time and should be avoided.

Subscribe and it lands in your inbox.

And if any of this sounds like where you are right now—stuck somewhere with a weird hobby that might be worth more than you think—tell me in the comments. I read all of them.

Follow @BowTiedCessna on X for more.

Need help getting hired in DevOps/SRE or want to chat about whether your background is a fit? Reach out at hello@bowtiedcessna.com