The largest contribution would be providing programming expertise as those doing science aren’t usually strong in that area. Making Linux and its tools more robust. Improving math and other libraries. Improving networking and data sharing. Even helping research groups develop more robust and reproducible code. The HN community, I assume, is skilled in all aspects of computerization it should be quite useful pushing science forward. If individuals are so furtunate, they can contribute money as well as (or rather than) effort.