Questions: Ants and rodents are recognized as ecosystem engineers worldwide creating soil‐modified nests and adjacent edges with superior resources. The influences of Dipodomys spectabilis (rat) and Pogonomyrmex rugosus (ant) on nests and edges would be similar in burned and unburned habitat on functional groups' cover, density, number of inflorescences and species number. I hypothesize that annual and perennial dicots would recover faster on rat patches than on burned grassland, dicots and grasses would recover faster on ant patches than on burned grassland, and annual dicots would recover faster on rat than ant patches, whereas grasses would recover faster on ant than on rat patches. Location: Semiarid grassland, Sevilleta National Wildlife Refuge, New Mexico, USA. Methods: I analyzed annual and perennial dicots, grasses and subshrubs in burned and unburned plots at 80 nests during cool‐ and warm‐seasons two years postfire. Data were collected at nest top (mound), edge, and 20 m (control) from the edge. Results: Cover, density and species richness of different functional groups were low on animal mounds in unburned and burned plots. Ant edges had greater perennial dicot cover, grass cover and richness while rats had lower grass density and richness than control at unburned and burned plots. Perennial dicot density and richness recovered faster at rat nests than control. Ant edges yielded faster recovery of grass inflorescences than control. Dicots recovered faster on rat than ant mounds whereas grasses grew faster on ant edges. Conclusions: Rats create nests enhancing density of recovering dicots and ant edges facilitate reproduction of recovering grasses. Equal regeneration between animal nest and grassland occurred commonly when fire reduced grass competition. Dysochory, granivory and fire intensity may influence recovery. Nests may be sources of reestablishment and seed dispersal. Recovery of grassland may be amplified as plants infill from engineered nests.