| Devops actually makes things a little easier I think. With Ansible, setting up a server on AWS feels a little like maintaining a vimrc file. Get a starter playbook from someone and run that to setup a server on AWS. Then you modify as needed as you go through your days as a developer. The great thing about Ansible playbooks is that it codifies the setup. You have a "recipe" that you can modify and run as needed. For me, this is much better than the days where I would set up a VPS "firing from the hip." If you had me setup 10 different servers, I would have set them up 10 different ways. I think for development knowledge in general, it might be good to look at the following people... http://www.kalzumeus.com/ (patio11 here) http://brennandunn.com/ http://unicornfree.com/ Take a read through. For these guys, it's all about products. In the end, that's all that matters. Maybe try for a change of perspective where you look at your skills in terms of products. In Amazon, there are probably projects where they lock developers in dark basements to learn arcane crafts so specialized that these engineers forget how to spell their name. These engineers would probably struggle just the same as you if the need for these specialties were to disappear. Faced with the need to "retool" - would they be able to do it? Some might, some might not. There is (and probably will be for quite some time) still a lot of need for people doing the sort of "LAMP" development that you cut your teeth on. Just because the stack wouldn't run an F-35 doesn't mean it isn't valuable. So, look at what you know in terms of business models rather than actual skills. Focus more on the product which people need and then fill in the skills. Do people need AWS and Devops? The client probably isn't going to fuss over this. The client just wants to ship a product. You handle the details. Most businesses start here and then they change as they get more successful. These projects still need the general guys to come in and get the first iteration shipped. Then later they might need to bring in more specialized engineers to deal with more specific issues. Eventually even the founders may hand over the reigns to "grown up" leadership. So, you need to figure out where your "market fit" is. You can't be everything to everyone (though at one time it seemed like we could be.) Once you find that, then just carve out some time to continually develop that and keep an eye out on where it's going. Maybe you could even have a side project which you could use as a plan B. You don't have to spend so much time on it that you are having to take away from other important things. Just plan a good hour of focus at certain regular intervals , even if it's just an hour every couple of weeks. You might find that it develops in an interesting direction and you may want to pursue it further. Further reading... http://letsworkshop.com/freelance-as-a-service/ |