We present a two-stage stochastic programming model for optimizing anesthesiologist schedules, explicitly accounting for uncertainty in surgery durations and anesthesiologist handoffs. To inform model design, we conducted an online survey at our partner institution to identify key factors affecting the quality of intraoperative anesthesiologist handoffs. Insights from the survey results are incorporated into the model, which optimizes the trade-offs between delaying anesthesiologist relief times, handoff costs, and under-staffing costs. To overcome the computational limitations of solving the extensive form of the proposed two-stage stochastic programming model, we develop a monolithic reformulation. Our computational experiments, based on real-world data from our partner institution, demonstrate that the reformulated model outperforms the extensive form in both computational times and optimality gap. Comparing our model’s outputs with current staffing practice at the partner institution shows that the proposed approach achieves substantial cost reductions. Finally, sensitivity analyses high- light the model’s ability to provide insights into trade-offs between handoffs, staffing levels, and relief times.