This paper presents DARA, the Driving Awareness and Reflection Assistant that makes drivers aware of potentially dangerous practices on how they hold the steering wheel, and helps them reflect. DARA utilizes a hand recognition component and a feedback one. The first recognizes how drivers hold the steering wheel and classifies their actions through a Leap Motion controller and machine learning. The second is comprised by a mobile application that provides drivers with feedback during and after their drive. DARA was evaluated with three studies for its accuracy, relevance and utility. Our findings show that DARA was successful both in making holding patterns present-at-hand for the drivers and in assisting them to reflect. We conclude our paper with a discussion on the broader implication of our findings on in-car hand recognition and feedback systems.