Chemical risk assessments typically focus on single substances, often overlooking real-world co-exposures to chemical mixtures. Mixture toxicology studies using representative mixtures can reveal potential chemical interactions, but these do not account for the unique chemical profiles that occur in the blood of diverse individuals. Here we used the H295R steroidogenesis assay to screen personalized mixtures of 24 persistent organic pollutants (POPs) for cytotoxicity and endocrine disruption. Each mixture was reconstructed at a human exposure relevant concentration (1×), as well as at 10- and 100-fold higher concentration (10×, 100×) by acoustic liquid handling based on measured blood concentrations in a Swedish cohort. Among the twelve mixtures tested, nine mixtures decreased the cell viability by 4–18%, primarily at the highest concentration. While the median and maximum mixtures based on the whole study population induced no measurable effects on steroidogenesis at any concentration, the personalized mixture from an individual with the lowest total POPs concentration was the only mixture that affected estradiol synthesis (35% increase at the 100× concentration). Mixtures reconstructed from blood levels of three different individuals stimulated testosterone synthesis at the 1× (11–15%) and 10× concentrations (12–16%), but not at the 100× concentration. This proof-of-principle personalized toxicity study illustrates that population-based representative chemical mixtures may not adequately account for the toxicological risks posed to individuals. It highlights the importance of testing a range of real-world mixtures at relevant concentrations to explore potential interactions and non-monotonic effects. Further toxicological studies of personalized contaminant mixtures could improve chemical risk assessment and advance the understanding of human health, as chemical exposome data become increasingly available.