[oelmekki]

For me, the answer is side projects. I keep playing with new ideas, and use new techs by the way.

It not only allows me to discover the tech, it's also especially important because I refuse to make first contact with a tech by implementing it directly in a production project meant to stay around for years. The most projects I've used a new tech in before introducing it in my main project, the more comfortable I am that I did not do gross mistakes.

For me, it's not the amount of time using a new tech that matters, it's the amount of projects I used it in (because each time, I can try a different architecture).