Do some ops stuff in your free time. Setup an awesome VPN with a jumpbox. Setup a MySQL/LAMP cluster. Put an nginx server in front of it. Replicate the database

Do some ops stuff in your free time. Setup an awesome VPN with a jumpbox. Setup a MySQL/LAMP cluster. Put an nginx server in front of it. Replicate the databases properly. Setup iptables properly and script the configurations so you can deploy new boxes. I did this for a while and it really helped to motivate me because it was different type of work, but it puts you in a position where literally all you have to do is write the code. You have a production stack, a launch pad so to speak, everythings nice and pretty, now all you have to do is build a rocket. Don't try to build a fantastic one on the first iteration, just do it. I find using Python for initial versions of anything is a good idea because it's quick to code. After that find the bottlenecks and micro-service using languages that seem appropriate for the various bottlenecks. Start small man. Don't focus on the light at the very end of the tunnel, you will think it's too far and get discouraged, especially if you're watching everyone else. Focus on the beta launch and the minimal set of features required. You'll be surprised after you see it running how inspiring it will be to start cranking out code.