The gender wage gap rises with experience. To what extent do firm policies mediate this rise? We use administrative data from Italy to identify workers' first jobs and compute wage growth over the next 5 years. We then decompose the contribution of first employers to the rise in the gender wage gap, taking account of maternity events affecting a third of female entrants. We find that idiosyncratic firm effects explain 20% of the variation in early career wage growth, and that the sorting of women to slower-growth firms accounts for a fifth of the gender growth gap. Women who have a child within 5 years of entering work have particularly slow wage growth, reflecting a maternity effect that is magnified by the excess sorting of mothers-to-be to slower-growth firms. Many entrants change jobs within their first 5 years and we find that the male-female difference in early career wage growth arises from gaps for both movers and stayers. The firm components in wage growth for stayers and movers are highly correlated, and contribute similar sorting penalties for women who stay or leave.