When COVID-19 swept the world at the end of 2019, it changed life as we knew it. With about 600 million positive cases (both recovered and active) and approximately 6.5 million deaths due to the disease, people worldwide have been affected physically, psychologically, economically, and socially by the pandemic. Amid such difficult times, @FacesofCovid—a Twitter account with more than 150,000 followers—was launched in March 2020 with the mission of honoring the lives of those lost to COVID-19 instead of presenting them as mere statistics. The account is a demonstrative example of the mourning genre as primarily exhibited through concise tweets grieving the deceased. As such, it offers a novel case of a public online mourning platform through microblogging, an understudied research area that merits further examination. A self-built corpus of 280,536 words was built from more than 7,000 tweets on the public account. The analysis presented in this paper focused on how people are constructed in the language of their loved ones as they are mourned through these tweets. Drawing on insight from van Leeuwen’s social actor representation and corpus linguistics, the analysis was conducted using the #LancsBox corpus processing software package. The findings indicated that gender asymmetry persists within this corpus. Therefore, this paper adds to the rich body of literature documenting gender imbalance across different genres and domains. Men are far more present than women and are constructed through functionalization for the most part, whereas women are less functionalized and represented primarily through relational identification. In light of this, it is argued that while sometimes, gender asymmetry can intentionally be ideologically loaded and may serve hidden agendas, at other times, it may inherently and subconsciously be passed on through spontaneous language use.