At the South Pole, in the middle of Antarctica, we have drilled deep holes down into the glacial ice to instrument a cubic kilometer of this highly transparent and pristine medium with very sensitive light sensors. This detector called “IceCube” allows us to capture the dim trace of light that some neutrinos leave behind when they interact in the ice. In order to use the collected raw data to answer scientific questions, a number of challenging data analysis problems need to be addressed. In this talk I will highlight some of the problems and solutions that I encounter in my personal research of neutrino oscillation physics. We will dive into the topics of event reconstruction and high statistics data analysis with small signals, discuss our current approaches and open issues.