3 ms·
Not sure if useful for anybody, but here is my story. I have no real cs-background, I just liked computers in high school and participated in some local program
by Zvez 6y ago
Not sure if useful for anybody, but here is my story.
I have no real cs-background, I just liked computers in high school and participated in some local programming contests.
I started working on my 4th year of uni. It was java enterprise. So yeah, a lot of crud. But even then we managed to build abstract stuff to make crud building easier and fun.
After around 6 years of crud I decided to try something different, and went to yandex (it is like google, but in Russia). First time it was just the same crud (lot of rest api), but after some time I managed to land on some infra projects - kafka, streams, microservices. And it became much much more fun. You really build stuff from the scratch, you design it, test it under high load.
Anyway my tips:
1. Don't go for buzz word, find job where you can actually choose technologies. Soon you will understand how to see through the hype. And cost of choosing wrong tech.
2. Stop trying to learn what 'everyone think is cool'. Learning compilers won't make you better developer, esp if you are not into it. And Dragon book is not the best intro to compilers anyway. If you want to learn something fun, find what it is fun for you in the first place.
3. If you want to become better developer, try to improve things you do at your day job. Not something like 'we should use k8s or we die'. More like 'we will have to make a lot of these crud in near future, lets spend some time to build framework that makes building and support them 2x faster and cheaper'. Look at small and achievable improvements first.
4. Invest time to understand how things work. And build your own opinion about technological decisions.
- jokab 6y ago> ...spend some time to build a framework... Yes, no. Do this and then leave the company. Then I or some other poor sucker gets to maintain your framework that makes you 2x faster but everybody else 2x slower. Yay fun. Build a framework only if its reeeeeaaaaallly necessary. Whis is most likely never. If you feel the urge to write one, by all means build one for your own side projects, but dont shove it down everybody else's throat