Managing and understanding the growing volumes of scientific data is one of the most challenging issues scientists face today. As analyses get more complex and large interdisciplinary groups need to work together, knowledge sharing becomes essential to support effective scientific data exploration. While science portals and visualization Web sites have provided a first step towards this goal, by aggregating data from different sources and providing a set of pre-designed analyses and visualizations, they have important limitations. Often, these sites are built manually and are not flexible enough to support the vast heterogeneity of data sources, analysis techniques, data products, and the needs of different user communities. In this paper we describe CrowdLabs, a system that adopts the model used by social Web sites, allowing users to share not only data but also computational pipelines. The shared repository opens up many new opportunities for knowledge sharing and re-use, exposing scientists to tasks that provide examples of sophisticated uses of algorithms they would not have access to otherwise. CrowdLabs combines a set of usable tools and a scalable infrastructure to provide a rich collaborative environment for scientists, taking into account the requirements of computational scientists, such as accessing high-performance computers and manipulating large amounts of data.