Religious divisions have long played a primary role in major conflicts throughout much of the world. Intergroup contact may increase trust between members of different religions. However, evidence on how inter-religious contact affects individuals' behavior towards one another is scarce. We examine this question in the setting of a four-year university in Lebanon, a country with a long history of deep divisions and armed conflicts between religious groups. To identify causal effects, we exploit the university's random assignment of first-year students to peer groups. We proxy students' religious backgrounds by whether they attended secular, Christian or Islamic high schools—the last of which have the most religiously homogeneous student body. Results indicate that for students from Islamic high schools, exposure to peers from different religious backgrounds decreases their enrollment in courses taught by instructors with distinctively Muslim names, suggesting that contact improves trust towards members of other religions. Moreover, we show that students from Islamic schools experience improvements in GPA when interacting with those from other groups, while exposure to Islamic students reduces secular students' academic performance.