The field of toxicology is experiencing a paradigm shift with high-throughput (HT) approaches that rapidly query huge swaths of chemico-structural space for bioactivity. Its practicality is due in large part to switching from high-cost, low-throughput mammalian models to faster and cheaper alternatives. We believe this is an improved approach because the immense breadth of the resulting data sets a foundation for predictive structure-activity-based toxicology. Moreover, rapidly uncovering structure-related bioactivity drives better decisions about where to commit resources to drill down to a mechanism, or pursue commercial leads. While hundreds of different in vitro toxicology assays can collectively serve as an alternative to mammalian animal model testing, far greater efficiency and ultimately more relevant data are obtained from whole integrated animal systems. The developmental zebrafish, with its well-documented advantages over many animal models, has emerged as a true biosensor of chemical activity. The potential of the developing zebrafish model for toxicology and drug discovery was limited by inefficient approaches such as, 1) plating zebrafish into multi-well plates; 2) chorion removal; 3) chemical exposures; 4) imaging; 5) adult behavioral assessments; 6) and data analysis. Over the years, we cleared these hurdles through instrument, protocol, and software developments to move zebrafish into the HTS environment.