3 ms·
The article touches a lot of interesting points, but often the novelty (and the shiny pay) is the main attraction point. For me, I just figured out that am goo
by n_ary 3y ago
The article touches a lot of interesting points, but often the novelty (and the shiny pay) is the main attraction point.
For me, I just figured out that am good at it and enjoy it when I was doing some embedded work for six months. Before that, I was doing a lot of matlab and wouldn’t dream of doing that as a profession ever.
Best to try it out a while as a small internship/apprenticeship(EU thing) before deciding if you want to do it long term.
I saw a very many people try to learn Python and then read cool things about C++/Rust folks and then try those to build basic CRUD and then loses interest. At the beginning best to use the easy tools and see if you can persist at it and once you have mastery, try to dip in more complex topics.
- zackproser 3y agoThanks for the reply! I agree with your point - most of the time when people find out what I do they ask about the comp first. I also think it's critical to have those progressive wins, and progressively more difficult, ambitious projects in order to keep learning and growing.