3 ms·
Here's advice from another almost 30 years old engineer :). So I might not have more experience, but here is how I think about things, maybe it helps. I think
by matijash 7y ago
Here's advice from another almost 30 years old engineer :). So I might not have more experience, but here is how I think about things, maybe it helps.
I think the most important question is: What interests you? And I believe the answer should be more general than a specific technology such as AWS or GCP.
You said you would like to grow as an engineer. There are typically 2 directions - going more into managerial/leadership roles (e.g. staff engineer) or being more of an individual contributor (no management, only engineering). Good book on this - Radical Candor.
It is also totally ok to switch between these roles.
About being afraid - what do you have to lose? With engineers being in such a high demand you definitely won't end up jobless. If you are thinking about doing it and are not happy with your current company, I'd say do it and fail as many times as needed until you get where you want to.
Another good way to learn and build a portfolio is doing a side project. You can then open-source it, write a blog post on it and sort of become an "expert" on that topic.
Regarding your wife and family, probably the best thing IMO would be to talk to them - explain them your position, ask for support and agree with them how it is going to work.
When I don't know what to do, what typically helps me is to take action. Any action. Once I start doing things, things start to unravel and become clearer very quickly.
Hope this helps!
- non-entity 7y agoSerious question: If you have 10 years experience in X, but is looking for jobs in Y, are employers, recruiters, etc. really going to care if all you have to show your competence is a few side projects and blog posts when the position wants 10 years of experience with Y?
- thrwlost 7y agothis was exactly my point, if you are lucky and joined company which has high potential to grow and/or joined Google like companies with crunching lots of competitive programming problems (leetcode, hackerrank) then you might join teams where people work on kubernetes, large scale storage, low latency proxies and so on. Even though some of them are open source, it is not easy to do contributions after whole day of coding. I did this in the past, to become productive in projects like k8s you should invest a lot of your own time. At small scale k8s rarely fails and features out of box is enough for most companies, since you don't have large scale problems/experience it is difficult to come up with feature which could be useful for k8s community. this applies to other projects as well. Unless you are working full time on field, it's not easy to become expert in that field