This paper studies the effects of gender interactions on the supply of and demand for credit using data from a large Albanian lender. We document that first-time borrowers assigned to officers of the opposite sex are less likely to return for a second loan. The effect is larger when officers have little prior exposure to borrowers of the other gender and when they have more discretion to act on their gender beliefs, as proxied by financial market competition and branch size. We also find that first-time borrowers matched with opposite-sex officers pay higher interest rates and receive smaller and shorter-maturity loans, but do not experience higher arrears. Our results are consistent with the existence of a gender bias and learning effects that lead to the disappearance of the bias.