Galactic-scale outflows power the `baryon cycle': they recycle gas from star-forming regions into the circumgalactic medium, and likely play a key role in determining the frequency of gas accretion (or re-accretion). However, the relationship between outflow kinematics and the star formation activity in distant galaxies remains poorly constrained. This deficit is a particular problem for theories of low-mass galaxy formation, as standard feedback prescriptions cannot reproduce the observed evolution in the star formation rates of these galaxies from z~7 to today. Observational constraints on cool gas accretion onto and outflows from these objects are needed to resolve this issue. First, I will discuss our recent work on cool gas outflows from normal, star-forming galaxies at z~0.5, showing that the observed outflows are indeed ubiquitious along the star-forming sequence and are physically capable of significantly affecting galaxy evolution. Then I will describe a new technique to measure the HI and metal-line absorption strength in low-mass galaxy halos at z~2, placing the first constraints on the incidence, spatial distribution, and kinematics of cool gas accretion and large-scale outflows in these systems.