python3 -c 'import pandas as pd; df=pd.read_csv("data.csv"); print(df.sort_values("salary", ascending=False).head(3))'

sort_values orders the DataFrame by the salary column, ascending=False puts the largest first, and head(3) keeps the top three rows. This is the pandas recipe for a top-N ranking. The default ascending=True gives the smallest values first. Chain reset_index() if you need fresh row indices after sorting.