Constraint-Based Local Search emerged in the last decade as a framework for declaratively expressing hard combinatorial optimization problems and solving them with local search techniques. It delivers tools to practitioners that enable them to quickly experiment with multiple models, heuristics, and meta-heuristics, focusing on their application rather than the delicate minutiae of producing a competitive implementation. At its heart, the declarative models are reminiscent of the modeling facilities familiar to constraint programming, while the underlying computational model heavily depends on incrementality. The net result is a platform capable of delivering competitive local search solutions at a fraction of the efforts needed with a conventional approach delivering model-and-run to local search users.