In 2018 alone, the US landfilled 35.3 million tons of food waste, about 24% of the total landfilled mass. In addition to the negative impacts landfills have demonstrated on the environment and human health, some states have begun to outlaw or dissuade the disposal of food waste and sewage sludge into landfills altogether. An urgent need has thus been created for the development of digestion processes like anaerobic digestion (AD) and arrested methanogenesis (AM) to convert food waste into valuable chemical products. Unfortunately, the buildup of volatile fatty acids (VFAs) during these processes eventually halts the reaction, and energy efficient methodologies for VFA removal are critical for the operation of fermenters. Additionally, VFAs themselves can serve as valuable chemical precursors, and recently AD processes have been modified to increase VFA production during fermentation. However, even with significant research over the past three decades, the separation of VFAs from the fermenter broth has remained expensive. Moreover, the separation of these VFAs from the fermenter broth may cost up to 50% of the entire process budget, hindering the widespread commercial adoption of AD and AM. In this work, we present a novel liquid–liquid extraction process termed CLEANS (Continuous Liquid-liquid Extraction And iN-situ Separation) as a highly efficient method for continuously separating VFAs from a real fermentation broth solely under gravity. Our optimized process (using an aqueous broth feed pH of 2.5, tri-n-octylamine as an extractant, and a 10:1 ratio of aqueous broth to organic extractant), achieved a VFA distribution constant KD = 44.5 ± 7.9, a single-pass recovery = 81.3 ± 2.5%, and an extraction factor = 8.1 ± 0.3. These KD values are over an order of magnitude higher than what has been previously reported for comparable processes. A high aqueous-to-organic flowrate ratio, enabled for the first time by CLEANS, was found to be particularly crucial for achieving optimal extraction. Our separation process demonstrates excellent reproducibility and potential for scalability. The economic and environmental implications of this work are briefly discussed.