Academics’ use of social media platforms is widely recognized and often understood as an extension of traditional academic practice. However, this understanding does not account for academics’ use of pseudonymous Twitter accounts. We used a combination of computational and human-driven methods to examine the activity of 59 anonymized, self-identified academics on Twitter. Our computational analysis identified five broad topics: discussing academic life, discussing British news and affairs, discussing everyday life, surviving lockdown, and engaging with academic Twitter. Within these broad topics, we identified 24 more specific codes, most of which were concentrated in individual topics, with some cross-cutting codes. These codes demonstrate how the pseudonymous accounts considered in this study can be considered ‘authentically academic’ even if they do not conform with widespread expectations of academic social media use.