that is $25K/year in income that no one pays taxes on

The hardest part in machine learning is that while we frequently use the same tools as software engineers, the thinking process is quite different. I work and have worked with some great software engineers as clients, and I hire software engineers as staff. And anyone can learn an API or figure out how to run Hadoop or Spark. This doesn't matter. The main differentiator is math. Projects live and die by the mathematical capacity of the engineers and data scientists working on them. In the end, machine learning is about taking some data, applying some math, and making a prediction. The coding is important--but it is secondary. A staff member who can do real math is worth 10X a good programmer. By that I mean they get their work done faster, require less managerial oversight, and produce better results. And they are not constantly trying to write fancy code. The most important thing you can do is brush up on your math skills. Basic calculus, linear algebra, and statistics.

Maybe no American wants to work for you I have a PhD in computational physics from an Ivy League school, and I was programming the worlds fastest supercomputers when you were learning how to memorize your multiplication tables back in India Being a good engineer is not about memorizing apis and answering coding questions--it's about knowing how to solve a problem you have NOT seen before ! This is why Indian univeristies produce code monkeys and the US Universities produce Nobel Lauereates.