I would suggest that you follow some of work done in conferences related to the field that you are interested in. See what topics are currently being tackled an

I would suggest that you follow some of work done in conferences related to the field that you are interested in. See what topics are currently being tackled and see if you can come up with initial hypothesis on the road to a solution. A PhD usually requires you to come out of "code-it-up" mode. You clearly have to state a sufficiently scoped problem, why it hasn't been solved before and why your solution can claim to solve it. If you are looking for more interesting jobs, it is easier just to switch companies. Many RnD companies don't exclusively hire PhD candidates. All the best :)