Secure Multi-Party Computation (MPC) enables a pair of millionaires to compute whose net worth is bigger, without having to reveal their net worth to anyone (and with the guarantee that the outcome would be correct even if the other millionaire tries to cheat). More generally, MPC allows multiple distrusting parties to jointly perform some computation on their private inputs, where the outcome of the computation is correct even in the presence of cheaters, and the inputs of the honest parties remain private. For example, such protocols enable multiple countries to pull together the medical records of their citizens to run machine learning algorithms on all records, to predict which will be the next main flu virus, what is the best treatment for high blood pressure, and much more. MPC is crucial in such scenarios, since the a large number of records is vital to obtaining accurate results, but privacy regulations prevent governments and organizations from sharing the records with entities from other countries. A recent proposal, developed together with Assistant Prof. Adi Akavia, Ben Galili, Dr. Hayim Shaul, and Prof. Zohar Yakhini, for privacy-preserving machine learning algorithms for COVID variants using fully-homomorphic encryption, won 3rd place in the 2nd (FHE) track of the 2021 iDASH competition.