Nowadays IT is present in our everyday life like: cars, phones, clothes, watches, houses etc. In this context the developing industry for such products is in a continuous need of specialists. One of the basic features of IT specialists is the ability to program, namely to be able to write code to be understood by machines and devices. Each year company representatives ask universities to double the number of graduates. Teaching students to write code is a complex and hard task proved by very high dropout rate after the first year of study in the distance learning programs. In order to ameliorate this issue and to help students learn the theoretical concepts of programming tutors can use efficiently learning objects deployed on LMSs. Learning objects can be simple or plain (LO), generative learning objects (GLO), and auto-generative learning objects (AGLO). GLOs offer instantiable reusable templates for different learning objectives. AGLOs add automatic instantiation enhanced with random variables in order to increase content diversity. The runtime of a program is for first year students like a black box, they do not control the code very well. Debuggers may be a solution in this sense, but usually they are built for experienced programmers, they have to be configured, sometimes they are difficult to follow, do not offer spectacular views of the changing data etc. In order to help students understand how their code behaves at runtime we propose an online code interpreter with runtime data visualizations and we analyze the impact of such a generative learning object on their progress. Thus, we consider that such efforts will contribute to the idea that the technology can support learning efficiency.