The MaTHiSiS learning vision is to provide a product-system for vocational training and mainstream education for both individuals with an intellectual disablity and non-diagnosed ones. This product-system consists of an integrated platform, along with a set of re-usable learning components (educational material, digital educational artefacts etc.), which will respond to the needs of a future educational framework, as drawn by the call, and provide capabilities for: i) adaptive learning, ii) automatic feedback, iii) automatic assessment of learner’s progress and behavioural state, iv) affective learning and v) game-based learning. To achieve these educational innovative goals, the MaTHiSiS project will introduce a novel methodology in the education process. The so-called learning graphs which, acting as a novel educational structural tool and associated with specific learning goals, will foster novel ways to guide how the different learning material and artefacts can be deployed throughout a prespecified learning scenario. The building materials of these graphs are drawn from a set of Smart Learning Atoms (SLAs) which will constitute the vertices of the graphs. SLAs are learning elements that carry stand-alone pieces of learning materials, targeting certain problems. More than one SLAs, working together on the same graph, will be able to help individuals reach their learning/training goals. The learning goals as well as the SLAs involved will be decided and pre-agreed based on common practices, goals derived from formal and non-formal education (general education, vocational training, lifelong training or specific skills learning) as well as learner’s own goals (so as to equally serve in-formal education contexts).