The AI frenzy is really making me rethink my career - What direction should I take?

As you get towards the more senior side of the ladder, the bias/preference for the expected impact actually leans towards problems than specific tools. Tools are simply a mechanic to solve problems, and don't provide any inherent value when used in isolation. Let's say I have 1000 hours: I could build a heavily polished AI-based dating app for cats or I could work for a fast food job for those hours. Generally, the fast food job will net me more value because the cat dating app provides no purpose or value to enough users where a growth/monetization plan can be executed.

If you feel like AI can solve the core problems you or your team face at work, then there's a clear problem space that can be addressed by becoming more of an AI engineer. Else, you're just giving into FOMO and wasting time that could be used to build up engineering fundamentals (communication, xfn leadership, domain knowledge, coding experience). Fundamentals will carry over with time, but anchoring yourself on specific tools or technology will not.

I have never understood the doomsday fearmongering around how AI is going to take over the world and every software engineer will have to be an MLE or unemployed. Every company with a powerful AI core still needs the following technical ICs:

• Backend engineers to package up the output of the model and expose it via APIs so others (businesses + consumer applications) can call it
• Frontend engineers to expose the output of the AI through some human-friendly interface (mobile, web, Mac/PC desktop app) - This is especially relevant if you're a consumer app, and even if you're a B2B SaaS company, you still need to expose a configuration/maintenance dashboard
• Data engineers and analysts to log the data behind the AI application usage and understand user behavior (plus identify bugs from data anomalies)
• Site reliability engineers to make sure that the servers stay up as with any big software system
• Security engineers to make sure that your APIs/AI don't get hacked and produce offensive results/leak data
• Internal tools engineers to make tools to train the models alongside all the other internal use-cases companies have (CI/CD, data visualization, recruiting). At Meta, we had thousands of engineers who just worked on internal tools (many of them very senior)
• Build engineers to make builds faster, more observable, and more reliable as you need to compile code into executable binaries

The list goes on and on and on. AI/ML is just a different way of building the business logic behind a software application and power a backend. Some people have blockchain, some will do vanilla old-school CRUD rules-based stuff, and some will use AI/ML. The fact that AI/ML is innovative/new doesn't mean it will usurp all other parts of the product tech stack.

At the end of the day, just do something you're passionate about and good at (and if you're truly passionate about something, you will almost certainly get good at it). If that's AI/ML, by all means, go for it. But if you're already doing web development or data engineering or something and you love it - Just stick with that.

I love mobile development, especially Android, which is why I've stuck with it for 10 years and built 30+ mobile side projects with 4 million+ users combined. Companies still offer me jobs that pay me tons of money (and will continue to do so), because I stuck with this one field and got extremely good at it. If you want career longevity with stellar compensation, it's all about depth. Pick something and become one of the best in the world at it.

Here's some other great resources to check out:

• "Is it worth transitioning to become a Machine Learning Engineer?"
• How To Find Your Path Among The Countless Paths In Tech
• "How do I move ahead in my career?"