Bioactive molecules produced by bacteria form the basis of our current antimicrobial clinical arsenal. Historically, the majority of antibiotic-producing bacteria have been isolated from soils. Yet, recent studies have shown that our traditional culture techniques are only capable of growing less than 1% of all bacteria in a given soil sample. Accessing even a small portion of the “unculturable majority” of microbes present in soils is likely to lead to the identification of new antibiotic producers and new antimicrobials. To attempt to access previously unculturable microbes and identify new antibiotic producers, we have used microbial diffusion chambers with Australian soil samples to cultivate bacteria in situ. Using this technique, we were able to increase recovery of bacteria 6-fold over standard methods. We generated a collection of 1250 microbes and assayed these for antibiotic activity against drug-sensitive and multidrug-resistant hospital pathogens. We found ~12% of our isolates were capable of inhibiting at least one of our test strains. By combining DNA sequencing and mass spectrometry approaches, we have dereplicated our antibiotic producing isolates, revealing several isolates producing potentially novel antimicrobials active against multidrug-resistant bacteria. Furthermore, investigating the phylogeny of these isolates showed that a diverse array of microbes had been cultivated, spanning the majority of the microbial family tree. This study shows there are still antibiotics awaiting discovery from soil bacteria and provides a glimpse into the antibiotic potential of Australian soil microbes.