I've been thinking the same. So far I studied a lot of the theory, and got familiar with using some of the most popular models. I'm thinking building a few toy

I've been thinking the same. So far I studied a lot of the theory, and got familiar with using some of the most popular models. I'm thinking building a few toy models will be the next step, and then maybe solving some specific problem as a demo. I've also done some interviews and got crushed but it lets me gauge what else I need. What I'm wondering is how much is needed to just be an engineer implementing the theory as oppose to a scientist researching the methods.