The birth of the sedentary communities in the Kura and Araxes Valleys at the beginning of the 6th millennium BC corresponds to a late Neolithisation process in Southern Caucasus compared to the Near East. Agriculture and animal husbandry appear to have been directly mastered from the beginning of the Neolithic and the discovery of a few North Mesopotamian ceramics at several sites, both in the Araxes and the Kura Valleys reinforces the hypothesis of relationships between the Caucasus and Near Eastern communities. Furthermore, recent genetic and paleo-genetic studies carried out on a set of faunal remains from Azerbaijan have allowed to trace the origin of goat domestication in the region, showing that those from Southern Caucasus would have come from Eastern Anatolia or North-West of Iran (Kadowaki et al., 2017). Also in genetics, studies carried out on the human genome have highlighted a genetic gradient (cline) from western Anatolia to the southern Caucasus resulting from a mixing process at work since at least the middle of the 7th millennium BC (Skourtanioti et al., 2020). However, while the mechanisms of this Neolithisation point to the adoption of an allochthonous system, there is also evidence of an endogenous development. The near absence of pottery at the earliest Neolithic sites (Haci Elamxanlı Tepe, Kiçik Tepe) and the similarity between the Neolithic (Haci Elamxanlı Tepe) and Mesolithic (Damjili Cave) lithic assemblages point towards a possible techno-cultural continuity with the local Mesolithic communities (Nishiaki et al., 2019). It is therefore possible that these contacts did not encourage a direct transfer of ceramic technology, and it seems more likely that Caucasian communities did not “imitate” exogenous techniques, but rather reappropriated a concept according to their needs and know-how (Palumbi et al., 2021). Similarly, we may argue that the development of architecture reveals the existence of original traditions in Southern Caucasus (Baudouin, 2019) compared to Mesopotamia where rectangular architecture has been known and mastered for several millennia (Baudouin, 2021). The birth of earthen architecture in Southern Caucasus ca. 6000 BC corresponds to Neolithisation process in the Araxes and Kura Valleys. In a general way, architectural practices evolved slowly corresponding to a strong “inertia” over more than 700 years. The circular shape, stretcher bond and the absence of quoin bonding are recurrent throughout the period, possibly a sign of a low level of technical knowledge of these communities. However, ca. 5700/5500 BC, changes in the organization of the household and a shift of domestic activities testify a deep socio-economic change in the Kura Valley. Thus, two different dynamics allow to distinguish two successive phases: a formative phase, between 6000 and 5700/5500 BC, marked by the establishment of architectural traditions, and a “second Neolithic”, between 5700/5500 and 5300 BC, where some architectural features develop probably linked to new needs within the community corresponding with another household organisation in the Kura Valley. The production of moulded mudbricks of plano-convex shape – flat on one side and domed on the other side – at Aruchlo and Mentesh Tepe (Azerbaijan) testifies to the technical mastery of architecture achieved by the communities of Southern Caucasus from the beginning of the Neolithic (ca. 5950 BC). Its appearance is concomitant with the establishment of the first sedentary societies in Southern Caucasus, which raises research questions about its origin: is it an autonomous innovation following the local development of innovative Mesolithic communities? Is it the result of a diffusion from Near East to Southern Caucasus of which the moulded mudbrick would be a component to be added to a Neolithic “package” already made up of agriculture and animal husbandry? Thanks to the analysis of the publications and excavation archives, observations on the field and micromorphological analyses made at Mentesh Tepe and to the contribution of ethnography and ethnoarchaeology, we can now reconstruct the stages of the chaîne opératoire of the first moulded mudbricks in Southern Caucasus. By cross-referencing archaeological data with a conceptual approach taking into account the phenomena of invention, diffusion, assimilation, borrowing and convergence, it is possible to understand the mechanisms underlying socio-cultural change within South Caucasus communities. This paper contributes to the current debate on the technical specificities developed by these communities with new data and thus to a better understanding of architecture, to define the complexities linking cultural “behaviours”, building materials and architectural solutions among these Neolithic communities. It is then necessary to return to the models already well established for Mesopotamia during the 7th-6th millennia (Sauvage, 1998), before proposing possible hypotheses for its appearance in Southern Caucasus. We have even been able to demonstrate that the standardisation of mudbrick sizes, combined with the adoption of a modular architecture in Mesopotamia, could be linked to the emergence of building specialists, thus providing new data on the social organisation of these communities and on learning processes during the Ubaid period. In Southern Caucasus, the technical disparities between Kura and Araxes Valleys would correspond to a local exchange network that reflects the particularities of the techno-cultural groups. The assumption is that populations with similar needs and technical capabilities in different micro-regional contexts will lead to a same technical response. The simplicity and relative homogeneity of techniques could be explained by the persistence of small groups where technical gestures are mastered by a part of the community. “Stylistic” variations in architecture should therefore be understood as regional dialects. In this context, the spread of the mudbrick mould from the Near East is problematic. Based on these results, we hypothesize that the production of Caucasian moulded mudbricks is – by its technical and morphological originality – a local and autonomous innovation specific to the Middle Kura Valley rather than the regional variant of a global technical diffusion from the Near East where the model of its diffusion is well known.