Among HIV, immune cell and drug, exhibit interactions that are usually not well understood and as a result, cannot be accurately modeled. In this paper, Modeling by AOC is to understand the dynamics of HIV infection and treatment. The use of AOC-by-self-discovery modeling was investigated. AOC-by-self-discovery methods try to adjust the system parameters automatically. To demonstrate the effects of therapies, we design and implement a HIV Computational Lab prototype. HIV Computational Lab is an AOC-based simulation of HIV immune dynamics that is currently being developed in NetLogo. It allows researches to investigate dependencies various immune responses to HIV. The HIV Computational Lab provides a good tool to characterize the process of HIV infection and study HIV drug treatment.