The Split State Model allows to obtain tamper and leakage resilience against a big class of attackers. In this model the secret material is divided in two halves that the attacker can only manipulate/leak independently. In the first part of the talk I will introduce the model, showing the peculiarities and the limits. Then I will introduce the concepts of Leakage-Resilient Codes (LRC) and Non-Malleable Codes (NMC) in the split-state model. In the second part of the talk I will introduce two new primitives which extend the notions of LRC and NMC and allows to get stronger form of security.