Struggling to explain why shoehorning AI into everything isn't the answer.

Question

I’m a master’s student in AI with 4.5 years of backend engineering experience. I’ve realized that making a cold-turkey switch to AI purely for higher pay is a flawed approach. Instead, I’ve decided to build deep backend expertise and soft skills in my current domain before transitioning into a related area like ML infrastructure or data pipelines.

My parents, however, are keen on me pivoting to AI sooner, likely because they started their careers in the 1990s and saw many developers’ skills become outdated during the rise of cloud computing and distributed systems in the late 2000s and early 2010s. This makes me wonder: are their incentives for me to 'futureproof' my career by chasing AI driven by the changes they witnessed during that earlier technological shift? If so, how can I balance their perspective with my longer-term career goals?

Rahul Pandey · Accepted Answer

IMO it's difficult to future-proof your career by self-learning a new framework or technology. This is especially true for something as broad as AI, where there are many different ways it will show up. I'm not saying you shouldn't chase AI, but I recommend you do it through a different lens:

Work at innovative companies, and work in areas adjacent to the innovation. For example, among Big Tech, it's pretty clear that Meta and Google are innovating, and in some ways creating the AI wave.
- So if you work at one of these companies for 5 years, the risk of you becoming redundant is minimal.
Work with innovative people who are trying out new ways of working. The best way to keep yourself updated is to work with people who are living in the future.

Finally, I wanted to mention that there's tremendous value in technical excellence in at least one domain of your career. If you are very valuable to your company as a backend engineer, you'll have plenty of opportunities to switch to other domains.

See this discussion: Breadth vs Depth as a junior engineer?

Struggling to explain why shoehorning AI into everything isn't the answer.

Discussion

Other Great Discussions