Resting-state functional connectivity (rsFC) measured with fMRI has been used to characterize functional brain maturation in typically and atypically developing children and adults. However, its reliability and utility for predicting development in infants and toddlers is less well understood. In Study 1, we use fMRI data from the Baby Connectome Project (BCP) study to measure the reliability and uniqueness of rsFC in infants and toddlers and predict age in this sample (8-to-26 months old; n = 170 sessions). We observed medium reliability for within-session infant rsFC in our sample, and found that individual infant and toddler’s connectomes were sufficiently distinct for successful functional connectome fingerprinting. Next, we trained and tested support vector regression models to predict age-at-scan with rsFC. Models successfully predicted novel infants’ age within ± 3.6 months error and a prediction R2 = .51. To characterize the anatomy of predictive networks, we grouped connections into 11 infant-specific resting-state functional networks defined in a data-driven manner. We found that connections between regions of the same network—i.e. within-network connections—predicted age significantly better than between-network connections. In Study 2, we applied the BCP-trained age prediction model from Study 1 to rsFC data from an independent sample of similar age range collected in India (INDIA; 8-to-26 months old; n = 107 sessions). We first assessed performance of this age-prediction model and found significant correlation between predicted and true age in the INDIA sample (Spearman ρ = .38, p < 1/500). Next, we asked if age-prediction errors (i.e., true age – age predicted from rsFC) were meaningful. That is, were model errors associated with developmentally relevant measures of brain structure and behavior? Specifically, we investigated relationship between age-prediction error and centiles from normative brain structure charts (Bethlehem, Seidlitz, et al, 2022) as well as parental reports of infant behavior measured with the culturally appropriate Ages and Stages Questionnaire (ASQ). Our hypothesis was that underestimations in age based on functional brain architecture (i.e., positive errors from the rsFC model) may be associated with other developmental-delay-like patterns such as lower centile in the brain structure charts or lower ASQ scores. Supporting our hypothesis, we found underestimated age predictions correspond to lower gray matter volume centiles (r = -.53 for cortical and r = -.56 ps < .001, for subcortical grey matter) across infants and toddlers. There was no significant correlation between the rsFC-based age prediction errors and white matter or ventricle volume centiles from the normative charts in this age range. In the behavioral domain, we found that underestimated age predictions may be associated with lower ASQ Problem Solving subscale (r = -.24, p = .044, N.S. after Bonferroni correction for the 5 tested subscales), which is in the hypothesized direction. Looking ahead, these findings can help characterize changes in functional brain organization in infancy and toddlerhood, inform work predicting developmental outcome measures in this age range, and bridge functional and structural brain measures of neurodevelopment.