In this talk I will present the SCIRun problem solving environment as applied to large-scale computational problems in science, engineering, and medicine. SCIRun is a parallel scientific programming environment that allows the interac tive construction, debugging and steering of large-scale scientific computations. Using this "computational workbench", a scientist can design and modify simula tions interactively via a visual programming environment. With SCIRun scien tists can automatically change parameters associated with a particular numerical method, boundary conditions or even the geometric structure of the computa tional mesh. As opposed to the typical "off-line" simulation mode - in which the scientist manually sets input parameters, computes results, visualizes the results via a separate visualization package, then starts again at the beginning - SCIRun "closes the loop" and allows interactive steering of the design, computation, and visualization phases of the simulation. I will provide examples of several driving large-scale applications of steering and interactive visualization in engineering (adaptive methods for CFD, resevoir simulation, and combustion) and medicine (cardiology, neuroscience, and imaging).