ImportanceFetal growth is precisely programmed and could be interrupted by environmental exposures during specific times during pregnancy. Insights on potential sensitive windows of air pollution exposure in association with birth weight are needed.ObjectiveTo examine the association of sensitive windows of ambient air pollution exposure with birth weight and heterogeneity by individual- and neighborhood-level stressors.Design, Setting, and ParticipantsData on a cohort of low-income Hispanic women with singleton term pregnancy were collected from 2015 to 2021 in the ongoing Maternal and Developmental Risks from Environmental and Social Stressors cohort in Los Angeles, California.ExposuresDaily ambient particulate matter with aerodynamic diameter less than 10 μm (PM₁₀) and aerodynamic diameter less than 2.5 μm (PM_2.5), nitrogen dioxide (NO₂), and 8-hour maximum ozone were assigned to residential locations. Weekly averages from 12 weeks before conception to 36 gestational weeks were calculated. Individual-level psychological stressor was measured by the Perceived Stress Scale. Neighborhood-level stressor was measured by the CalEnviroScreen 4.0.Main Outcomes and MeasuresSex-specific birth weight for gestational age z score (BWZ). The associations between air pollutant and BWZ were estimated using distributed lag models to identify sensitive windows of exposure, adjusting for maternal and meteorologic factors. We stratified the analyses by Perceived Stress Scale and CalEnviroScreen 4.0. We converted the effect size estimation in BWZ to grams to facilitate interpretation.ResultsThe study included 628 pregnant women (mean [SD] age, 28.18 [5.92] years) and their newborns (mean [SD] BWZ, −0.08 [1.03]). On average, an interquartile range (IQR) increase in PM_2.5 exposure during 4 to 22 gestational weeks was associated with a −9.5 g (95% CI, −10.4 to −8.6 g) change in birth weight. In stratified models, PM_2.5 from 4 to 24 gestational weeks was associated with a −34.0 g (95% CI, −35.7 to −32.4 g) change in birth weight and PM₁₀ from 9 to 14 gestational weeks was associated with a −39.4 g (95% CI, −45.4 to −33.4) change in birth weight in the subgroup with high Perceived Stress Scale and high CalEnviroScreen 4.0 scores. In this same group, NO₂ from 9 to 14 gestational weeks was associated with a −40.4 g (95% CI, −47.4 to −33.3 g) change in birth weight and, from 33 to 36 gestational weeks, a −117.6 g (95% CI, −125.3 to −83.7 g) change in birth weight. Generally, there were no significant preconception windows for any air pollutants or ozone exposure with birth weight.Conclusions and RelevanceIn this cohort study, early pregnancy to midpregnancy exposures to PM_2.5, PM₁₀, and NO₂ were associated with lower birth weight, particularly for mothers experiencing higher perceived stress and living in a neighborhood with a high level of stressors from environmental pollution.