Some effort has been employed to allow interpreted languages to be able to take advantage of the computing capabilities of GPUs. This makes sense, because it abstracts the hardware and its specificities away from the user application, making development less complex. However, due to hardware dependencies, the code needs to be compiled prior to execution. We want to compile the Lua function code into a GPU kernel as transparently as possible, allowing the user to access the underlying hardware, without the complexities related to the traditional GPU programming. This scenario presents a great challenge on how to infer variable data types with as little interference as possible on the user programming paradigm.