Counting Amount Of Columns With Different Content Under Number Of Conditions?

I've been desperately trying to figure out a formula that allows me to count the number of rows in a column that have different values, but that only counts those rows in which other conditions have been met. Probably sounds a little confusing, which is why I'll illustrate it with a screenshot:

The file contains a listing of major executives of a variety of different firms in different years from 1992 onwards. In the screenshot for example, you see the executives of AAR Corp from 1992 to 1997. Since my file however consists of 240,000 rows, there's a number of c. 2400 firms with their executives over a time period of 20 years each. Since not necessarily all executives remain with their companies for the rest of their lives, some executives show up for a number of different firms in my sample.

I want to determine at how many firms each executive worked within his current industry up to the respective year, based on the information of the 2400 companies and executive data over 20 years. The current industry is classified by 1 digit SIC codes (column U) and each firm has an individual firm identifier which is the CUSIP in column S.

For example, when applying this to the first row in the screenshot (Ira Eichner), I would need a formula that counts the number of different CUSIPs (thus firms), given that the EXECID (executive ID, identifier for each executive; column I) equals 9248, the 1 digit SIC code (column U) equals 5, and the Year (column J) is below 1992.