This as-told-to essay is based on a conversation with Manoj Tumu, a 23-year-old machine learning engineer at Meta based in Menlo Park. It’s been edited for length and clarity. Business Insider has verified Tumu’s salary and employment.
Machine learning has gone mainstream.
I started my master’s program in 2022, around the time ChatGPT was released. The advancement of AI and AI tools has made it a very competitive field, and many people are trying to get in.
It took me one year to get my undergraduate degree because I had college credits from classes I took in high school. Then I worked full-time as an engineer while doing my master’s in AI. After my master’s, I landed a role at Amazon, where I worked for nine months as a machine learning software engineer.
While I was at Amazon, I saw Meta doing a ton of cool machine learning things. When interesting roles came up, I just applied on their website or LinkedIn. Though I had learned a lot at Amazon, I just thought there was more interesting work going on at Meta.
In June, I left Amazon to join Meta as a machine learning software engineer for a total compensation of over $400,000. I was really excited about it and knew I wanted to take the job as soon as I got the offer. Here’s my advice for trying to break into this field.
Understand machine learning and its shifts
One of the things about machine learning roles is that there’s a lot of variation in the title. Depending on the company, it could be a research scientist, applied scientist, software engineer, or machine learning engineer. At Meta, I’m a machine learning software engineer on an advertising research team.
My role is a mix of research and implementation, but more research. With all the advances in AI, there are a lot more papers on AI and machine learning to read. I focus on ensuring that Meta uses the latest research and models to stay on the cutting edge.
Related stories
Business Insider tells the innovative stories you want to know
Business Insider tells the innovative stories you want to know
Machine learning has shifted so much. It used to be a lot more acceptable to just use classical techniques, which rely on humans to make decisions about data representations. Now the focus is on deep learning, which taps into artificial neural networks to automatically learn features from raw data.
People really see how powerful it is, which has resulted in non-tech companies investing in machine learning systems.
Try to get a tech internship while in college
Most of my applications to Big Tech companies have been cold applications on the website. I didn’t get a referral for my role at Amazon or Meta. However, I felt it was pretty easy to get an interview, since I had a decent résumé.
Experience is the biggest factor. In college, try to get an internship of any kind. I think those stand out the most. When I see the résumés that people post online, asking for advice, I see projects or programming languages taking up space.
Highlight experience over projects on your tech résumé
Highlighting projects you’ve worked on is good, but I think they’re overemphasized on résumés. They’re useful to include, but should be more supplemental. By the time I started applying for Amazon and Meta roles, I had already removed my projects from my résumé.
My general advice would be that once you have two or three years of experience, it’s OK to remove the projects and focus more on highlighting your experience.
Big Tech has a very standardized interview process — avoid one common mistake
The Meta recruiter emailed me and said they came across my profile, but never told me how. Soon after, we set up a call, and I reapplied.
The process was similar to when I interviewed at Amazon: a first screening, and then I moved through four to six interview rounds where they asked coding, machine learning, and behavioral questions. It took about a month and a half and was one of the smoothest interview processes I’ve had. I was very happy with everything.
One mistake I think people make is winging it during the behavioral interview, which thankfully I didn’t do. I studied the company’s values to prepare. I had a huge document where I wrote down stories to answer each question and follow-ups I would have.
Amazon, for example, has these leadership principles, and I studied those online and prepared stories tailored to them when I interviewed there. Meta has some that you can study, which it posts on its website, and I tailored my stories specifically to those.
Don’t worry about pay when you’re first entering machine learning
I made the mistake of not doing an internship while I was in college. But I was able to secure a contract role right after undergrad. This was at the start of 2022, when it still would’ve been possible to get a software engineering role without much experience.
Apply for any internship, even the low-paying ones that may not seem great. When I was choosing between machine learning roles and software engineering roles for my first job, I decided I wasn’t as interested in the traditional software engineer roles.
I chose a lower-paying machine learning role before I started working at Amazon, which I think really opened up more doors for me later.
Do you have an AI career story to share? Contact this reporter, Agnes Applegate, at aapplegate@businessinsider.com.
