The study aims to examine whether and how digital activities influence sleep issues among working Chinese youth. Methods: This study used data from the China Family Panel Studies (CFPS) a Chinese population-based survey, and employed the multilevel ordinal logistic regression model (MOLRM) to test the associations between digital engagement (whether to engage in digital activity, frequency, and duration) and sleep issues (bedtime, sleep duration, and quality) among Chinese working youth. Additionally, the restricted cubic spline model (RCSM) was adopted to fit the MOLRM to evaluate the nonlinear relationship between digital activity duration and sleep quality, and thus determine the optimal range of digital activity duration. Results: The analysis included 7849 working young adults. The digital usage rate was 84.11%. Digital use was not significantly associated with average, workday, or free-day sleep duration, after controlling for all potential confounders. However, most digital activity indicators could significantly predict bedtime and sleep quality. Furthermore, the RCSM indicated a non-linear relationship pattern between digital activity duration and sleep quality, with a weekly peak point of 25 h. Age significantly moderated the relationship between digital activity, sleep duration and bedtime. Younger youth who used digital media more frequently and for a longer time tended to sleep later and had shorter sleep duration than older youth. Conclusion: Digital usage significantly predicted later bedtime among Chinese working youth; however, it was not linked with sleep duration on workdays or free days. In parallel, a nonlinear correlation between digital activity duration and sleep quality indicated that appropriate digital activity duration (less than 25 h weekly) may contribute to good sleep quality.