Computer and smartphone-based applications for second language (L2) learning have become popular tools, being integrated in many classroom-based courses and adopted by the public at large. Yet, despite a significant body of research that suggests that individuals differ in their ability to learn L2, it is still unclear what factors predict successful L2 acquisition and how L2 teaching software can be designed to adapt to individuals' strengths and weaknesses. Here, we describe the architecture of LANGA, an online game-based platform under development for L2 teaching and research, and present a demonstrative proof-of-concept study using the platform. LANGA is designed to be both an effective and engaging product from the consumer perspective, and a tool that can be used by researchers to easily implement, deploy and test different training modalities for L2 teaching. Furthermore, key features of LANGA include easy configuration of training via modular design; emphasis on gamified teaching methods; and the use of automated speech recognition to provide learners feedback on verbal production. A first prototype of LANGA was tested in a small-scale, proof-of-concept study. Changes in proficiency from pre- to post-training were measured using recall and recognition tests, while event-related brain potentials (ERPs) were used to assess changes in brain activity related to lexical access over the course of learning. The results provided initial validation of the platform: participants were able to learn a large proportion of the words taught, and retained the novel words in a two/weeks follow-up. Future directions on the development of the platform are discussed.