The interaction engine hypothesis postulates a uniquely human ability and motivation for social interaction. A crucial juncture in the ontogeny of the interaction engine could be around 2-4 years of age, but observational studies of children in natural contexts are lacking. Moreover, data on the ontogenetic development of joint action coordination can provide critical human data for comparative research investigating the phylogenetic roots of the interaction engine. In this study we report on focal observations on 31 children aged 2 and 4 years old in 4 preschools (approximately 20h per child). Interactions were videotaped, transcribed and coded for the presence and duration of joint actions, their type (e.g., play, conflict) and the partners involved, as well as the presence and duration of entry, exit and main body phases of those joint actions. Children interact with a wide range of partners, many infrequently, but with one or two close friends. 4 year olds engage in cooperative joint actions more often than 2 year olds, and fight less than 2 year olds. Conversations and playing with objects are the most frequent action types. Joint actions are relatively brief (28s on average) and shorter than those of great apes in comparable studies. Their joint actions feature entry and exit phases about two-thirds of the time, less frequently than great apes. However, children interact with many more partners than great apes do, about 20% of their joint actions are multiparty and overall they engage in joint actions more frequently than great apes (on average 13 distinct joint actions per hour). The results support the interaction engine hypothesis, suggesting that, at a very young age humans already manifest a remarkable motivation and ability for fast-paced, flexible social interaction with multiple partners.