In this randomized evaluation, we study the effects of innovation contests (hackathons) organized within the state bureaucracy on civil servants' motivation, job satisfaction, their perceptions of the work environment and organizational culture. We implement a cluster-randomised design, implying that entire organisational units will be randomized into a treatment or a control group. For this pilot study, we collaborate with a local government in Uruguay and randomise the intervention across 65 organisational units. In the treatment group, civil servants will participate in a 2-3 month long innovation contest. In this contest, they will be invited to develop innovative suggestions for policy reforms to address a challenge defined by the local government.