The engineering of biological systems with predictable behavior is a challenging problem. One reason for this difficulty is that engineered biological systems are embedded within complex and variable host cells. To help enable the future engineering of biological systems, we are studying and optimizing the interface between an engineered biological system and its host cell or ``chassis''. Other engineering disciplines use modularity to make interacting systems interchangeable and to insulate one system from another. Engineered biological systems are more likely to work as predicted if system function is decoupled from the state of the host cell. Also, specifying and standardizing the interfaces between a system and the chassis will allow systems to be engineered independent of chassis and allow systems to be interchanged between different chassis. To this end, we have assembled orthogonal transcription and translation systems employing dedicated machinery, independent from the equivalent host cell machinery. In parallel, we are developing test systems and metrics to measure the interactions between an engineered system and its chassis. Lastly, we are exploring methods to``port'' a simple engineered system from a prokaryotic to a eukaryotic organism so that the system can function in both organisms. / Poster presented at the 2005 ICSB meeting, held at Harvard Medical School in Boston, MA.