To dissect cell-type composition and visualize cell clusters in space, we designed a panel of genes identified by dissociated single-cell transcriptomics according to their specificity and expression. Multiple regions of fresh frozen human fetal brain samples from gestational week 20 were sectioned to a glass coverslip, assembled into the platform’s hardware, and automatically probed for 30 genes using on-system cyclic smFISH chemistry. A full experiment from flow cell assembly to the probing of 30 genes in 10 cycles was completed within two days for a whole human fetal brain section without intervention. Single mRNA molecules across all levels of expression were consistently detected with high signal strength, sensitivity, and specificity using built-in computer vision algorithms. The results obtained with this robust, automated platform enable validation of functionally distinct cell populations identified by scRNA-seq and further reveal the spatial context of developmental gene expression programs at the single cell level. Taken together, our data enable direct observation of area-specific molecular diversity and the ability to infer functional roles of regionally diverse clusters based on known differentiation trajectories and migratory gradients.