Use of human pluripotent stem cells (hPSCs) has an unique potential for cell therapy since they could allow the production of almost any cell type at a large scale. Accordingly, many protocols have been developed to differentiate hPSCs into cells of clinical interest, including hepatocytes. However, these protocols are neither efficient nor robust and result in immature hepatocytes that can perform only some functions of adult liver cells. To optimise such protocols, it is imperative to understand the signalling pathways directing liver development. However, such knowledge is difficult to capture especially in human for obvious ethical reasons. As an alternative to primary tissues, our lab recently established a method to derive and grow primary human hepatoblast organoids (HBOs) isolated from human foetal liver tissues. These cells accurately and reproducibly recapitulate the hepatoblast identity in vitro, as they possess self-renewal and differentiation capacities, like the in vivo counterpart. Importantly, hepatoblasts are the first stem cell type of the liver and thus, they provide a new platform to study liver organogenesis. In this dissertation, the aim is two-fold. Firstly, to exploit the properties of HBOs for dissecting the signalling pathways that contribute to hepatoblast identity and secondly, to use this information to optimise a protocol for the generation of bona fide hepatoblast organoids in vitro from hPSCs. For the first part, I describe the effect of Wnt signalling pathway on hepatoblasts, showing that absence of Wnt cues results in diminished self-renewal capacity without directly inducing differentiation. Also, I demonstrate that Wnt presence can reactivate proliferation during the hepatoblast-hepatocyte transition. On the second aim, I develop an alternative differentiation protocol for the generation of an hPSC-derived hepatoblast cell population. To that end, I successfully generated a self-renewing pool of hepatoblast-like cells that display markers and characteristics similar to primary HBOs. Overall, the research presented in this dissertation has advanced the understanding of human liver development. Moreover, the generation of hepatoblast cells using our optimised protocol could provide a new source of liver cells for disease modelling and regenerative medicine.