Abstract: The stick-slip type of dynamics occurs in many natural processes including various types of avalanches and earthquakes. The stick-slip regime is characterized by periods of slow dynamics during the slip regime interspersed by abrupt changes know as slip events. Despite the importance of this problem and the active research, the mechanisms leading to granular failure are still not properly understood. In particular, predicting the slip events with high accuracy is one of the great challenges. In this presentation, we will discuss how topological data analysis can be applied to obtain new insights into the dynamics of sheared granular systems exhibiting stick-slip regimes. We will demonstrate that topological methods can be used to describe and quantify complicated patterns exhibited by force networks in a simple but meaningful way. We will use these techniques to analyze the dynamics of force networks in granular material in stick-slip regime. We will demonstrate that the evolution of force networks provides relevant precursors for the slip events.