Gender inequality in scientific careers can emerge at two analytically distinct margins: inequality in who remains in science and inequality among those who remain. While the former has been examined in research on attrition in science, much less is known about gender disparities among scientists who remain active for a long period. We address this gap by following survivors in science - scientists who started publishing in 2000 and continued publishing until 2020-2023 (N = 41,424). Together, they authored more than 2 million publications (N = 2,089,097) containing over 70 million cited references (N = 73,118,395) and worked in 38 OECD countries. Using raw Scopus data, we examine gender disparities in publishing intensity, international collaboration, journal prestige, productivity, citation impact and team size. Our findings reveal that men were on average 23% more productive than women cumulatively in 2000-2023 and 19% more productive in 2019-2023. By contrast, women and men published in similarly prestigious journals, achieved similar field-normalized citation impact, and worked in similarly sized teams. The structure of gender inequality is career-stage dependent: Disparities observed in broader populations of scientists need not persist among those who remain active in science longer, while the gender productivity gap persists.

翻译：暂无翻译