The association between suboptimal infant feeding practices and growth faltering is well-established. However, most of this evidence comes from cross-sectional studies. To prospectively assess the association between suboptimal infant feeding practices and growth faltering, we interviewed pregnant women at 28–32 weeks’ gestation and followed-up their offspring at postnatal months 3, 9, 16 and 24 months in rural Bangladesh. Using maternal recall over the past 24 hours, exclusive breastfeeding (EBF) status at 3 months, age at complementary feeding (CF) initiation, and receipt of minimum acceptable diet (MAD; as defined by WHO) at 9 months were assessed. Infant length and weight measurements were used to produce length-for-age (LAZ) and weight-for-length (WLZ) z-scores at each follow-up. Generalized estimating equations were used to estimate associations of LAZ and WLZ with infant feeding practices. All models were adjusted for baseline SES, infant sex, maternal height, age, literacy and parity. Follow-up was completed by 2189, 2074, 1969 and 1885 mother-child dyads at 3, 9, 16 and 24 months, respectively. Stunting prevalence increased from 28% to 57% between infant age 3 and 24 months. EBF at 3 months and age at CF initiation were not associated with linear infant growth, but receipt of MAD at 9 months was. By age 24 months, infants receiving MAD had attained a higher LAZ compared to infants who did not receive MAD (adjusted β = 0.25, 95% CI: 0.13–0.37). Although prevalence of stunting was already high at age 3 months, ensuring infants receive a diverse, high quality diet from 6 months onwards may reduce rates of stunting in the second year of life.