Acute myeloid leukemia (AML) is an aggressive hematological malignancy with a dismal survival rate, highlighting the need for therapeutic intervention. A major hallmark of AML is the failure to terminally differentiate into functional mature myeloid cells. Therefore, strategies to overcome this differentiation blockade represent an alternative therapeutic approach for treating AML patients. However, to date, there has been a lack of genome-wide assessment of genes involved in maintaining the undifferentiated state of leukemia cells. Here, I have adopted an integrated approach by combining loss-of-function CRISPR/Cas9 genome-wide screen with a cell-surface-based readout, which identified a number of novel, as well as known negative regulators of AML differentiation. Our top candidate was the RNAbinding protein (RBP), ZFP36L2, which demonstrated a highly selective “non-oncogenic” dependency in AML but dispensable for other cancer cell types. Mechanistically, ZFP36L2 physically associates with the 3’ untranslated region (3’ UTR) of several key myeloid maturation genes promoting their mRNA degradation and blocking myeloid maturation. Genetic inhibition of ZFP36L2 restores mRNA stability of these transcripts and ultimately triggers leukemic cells to undergo monocytic differentiation. Furthermore, epigenome profiling of a number of primary AML patient samples revealed distinct enhancer modules nearby ZFP36L2 that associated with distinct AML cell states, establishing a coordinated epigenetic and post-transcriptional mechanism that shapes AML differentiation. Taken together, this study provides a framework of genes that play a pivotal role in leukemic maturation and establishes the RNA-binding protein, ZFP36L2 as a key post-transcriptional regulator of AML differentiation. RBPs regulate many aspects of co- and post-transcriptional processes and, as such, aberrant expression of RBPs has been linked to tumorigenesis. Based on my findings and others, RBPs have been shown to exhibit cancer-specific dependencies and therefore have therapeutic implications. However, despite recent progress in uncovering RBPs required for tumor maintenance, the promise of targeting RBPs therapeutically is still limited by a lack of systematic evaluation for required RBPs in cancer. To systematically identify “non-oncogenic addiction” of RBPs, I have performed a CRISPR/Cas9 domain-based, loss-of-function screen targeting approximately 490 canonical RBPs across a wide-range of cancer types. My screening approach led to the identification of several RBP dependencies in AML, including candidates such as ZFP36L2 that are exclusively required for AML cells. Our recent work highlights that the splicing factor, RBM39 physically interacted with a network of RBPs upregulated in AML patients. We demonstrate that genetic or pharmacologic depletion of RBM39 in vitro and in vivo AML models, led to aberrant splicing of multiple members of its interacting RBP network as well as of transcriptional regulators required for AML survival. The effects of RBM39 loss on alteration of splicing further resulted in preferential lethality of AML cells bearing spliceosomal gene mutations, thereby providing a novel strategy for treating AML patients bearing RBP splicing mutations. Overall, my work demonstrates the role of RBPs in AML maintenance and provides insight into understanding the basis for their cancer-specific functions, which ultimately may represent a new avenue for therapeutic targets to improve clinical outcomes in patients.