One of the more promising routes that I've seen working well here is people who have put themselves inside organisations with large budgets and helped decide where that money goes. For example, if you are concerned with global poverty you could become a programme manager at the World Bank and quite plausably move $100m to more effective causes. If you cared about x-risk, an option would be to work for DARPA or IARPA and move $10m's to more effective research, and to help prioritise the research so that technolgoies are developed in an order that we think is less likely to cause x-risk. Another example would be to locate yourself within a large grant-making foundation.

The big downside with this approach is that the funds are usually less fungable than personal funds. How to weigh this against earning to give depends on your beliefs on the relative value of different activities that you would or wouldn't be able to fund. Through this approach you are typically able to control larger amounts of money than you would through earning to give.