My career path to Google included bartending and working at a nursing home. Here's what it taught me.

• Milica Cvetkovic, 35, took a non-traditional path to working in AI at Google.
• After college, she worked as a shot girl on Bourbon Street and then worked at a nursing home.
• She compares her career to a marathon, and said her combined experiences led to her current role.

This as-told-to essay is based on a conversation with Milica Cvetkovic, a 35-year-old senior technical solutions consultant at Google, based in Chicago. Her identity and employment have been verified by Business Insider. The following has been edited for length and clarity.

I've been working on building AI systems for Google Cloud customers for about three years, but my path to get there was a little non-traditional.

I'm not one of those people who applied to Google 100 times. I took a lot of different routes along the way. My career journey has been wild.

I improvised; I had moments when I was poor without health insurance; I had moments when I was lonely and helpless; I had moments when I messed up and was scared. But I also met the most magical people and had the best adventures. Regardless of how hard it was, I would change very little.

I was born and raised in Serbia, and moved to the US when I was 18. Instead of going to law school in Serbia, I took a gap year, bartended, and then went to college and majored in math.

After college, I moved to downtown Chicago and became a bartender. I love bartending. If I had time, I would do it now. Then I moved to New Orleans and worked as a shot girl on Bourbon Street for a while.

Next, I moved to Columbus and worked as a caregiver at a nursing home, which was probably one of the lowest moments of my life. I moved to Columbus, where I knew nobody, and where it was hard to meet new people. Working in the nursing home exposed me to seeing people die for the first time.

These experiences taught me that discipline matters a lot. When people say you have to work hard, that means you have to have discipline. I am not one of those people who were brought up with this hardcore discipline — I had to build it myself.

Eventually, I found a role somewhat related to what I studied and became a curriculum designer at McGraw-Hill Education for four or five years. At some point, I visited friends in Madison, Wisconsin, and decided I wanted to live there.

So I moved there and started grad school for statistics at the University of Wisconsin-Madison. The whole time I was in grad school, I continued working remotely for McGraw-Hill Education.

I started teaching

In grad school, I began researching how to teach machine learning to nontechnical audiences, specifically computational biologists. That experience, combined with my education in curriculum design, opened a lot of doors for me in education.

During my last semester of grad school, I started an internship as a machine learning engineer with a local startup in Madison that was using a device to predict animal health. I ended up working there full-time after grad school and also started teaching, first in bootcamps and eventually at the university level.

At some point, I didn't want to code 24/7 anymore. I felt like my soft skills weren't being utilized enough. Around that time, I saw that Google had posted my current job, and it was a perfect fit.

I think it's very valuable to have the skill to talk about technical concepts in a way that people are going to understand. The most incredible talks I've heard from accomplished scientists are from those who can talk about their research in an interesting and non-technical way.

You don't have to kill people with terminology and keywords to make a point. A lot of times when people do that, their audience gets confused. So I think having that ability is probably the most valuable skill that I bring.

Getting the job was like training for a marathon

I had failures along the way. It's not like everything went smoothly. I'm an immigrant, and for most of the time, I was alone in all of this. I've been through some really stressful situations.

When I started teaching at the university level, I worked very hard to get there, and I finally did, but I was super sick. Instead of being upfront about that, I ended up in the ER twice during that class, and I didn't deliver the way I wanted.

You have to move on in those moments. When you fail, you do damage control, and if the damage control doesn't work, you move on and do better next time.

People come from different backgrounds and situations. You don't have to hustle and have 17 different things you do in order to get a Big Tech job. You can be excellent in just one thing; it just turns out I had all those interests.

There wasn't one experience that made me get the job — it was everything. I worked in education; I was an engineer; I was technical; I had the degrees that mattered; I could speak to non-technical audiences; I had work experience; and I was active in the tech community.

As someone who ran a marathon, I can say that getting the job at Google was like training for the marathon. The marathon itself is more of a celebration of all the work that you've already done. That's what my application to Google was.