Hydrogels are widely used as three-dimensional (3D) tissue engineering scaffolds due to their tissue-like water content, as well as their tunable physical and chemical properties. Hydrogel-based scaffolds are generally associated with nanoscale porosity, whereas macroporosity is highly desirable to facilitate nutrient transfer, vascularization, cell proliferation and matrix deposition. Diverse techniques have been developed for introducing macroporosity into hydrogel-based scaffolds. However, most of these methods involve harsh fabrication conditions that are not cell friendly, result in spherical pore structure, and are not amenable for dynamic pore formation. Human tissues contain abundant microchannel-like structures, such as microvascular network and nerve bundles, yet fabricating hydrogels containing microchannel-like pore structures remains a great challenge. To overcome these limitations, here we aim to develop a facile, cell-friendly method for engineering hydrogels with microchannel-like porosity using stimuli-responsive microfibers as porogens. Microfibers with sizes ranging 150-200 μm were fabricated using a coaxial flow of alginate and calcium chloride solution. Microfibers containing human embryonic kidney (HEK) cells were encapsulated within a 3D gelatin hydrogel, and then exposed to ethylenediaminetetraacetic acid (EDTA) solution at varying doses and duration. Scanning electron microscopy confirmed effective dissolution of alginate microfibers after EDTA treatment, leaving well-defined, interconnected microchannel structures within the 3D hydrogels. Upon release from the alginate fibers, HEK cells showed high viability and enhanced colony formation along the luminal surfaces of the microchannels. In contrast, HEK cells in non-EDTA treated control exhibited isolated cells, which remained entrapped in alginate microfibers. Together, our results showed a facile, cell-friendly process for dynamic microchannel formation within hydrogels, which may simultaneously release cells in 3D hydrogels in a spatiotemporally controlled manner. This platform may be adapted to include other cell-friendly stimuli for porogen removal, such as Matrix metalloproteinase-sensitive peptides or photodegradable gels. While we used HEK cells in this study as proof of principle, the concept described in this study may also be used for releasing clinically relevant cell types, such as smooth muscle and endothelial cells that are useful for repairing tissues involving tubular structures.