The systems engineering aspects of evolving and developing the optimal design for Fourier transform interferometers are presented in this paper. A Fourier transform spectrometer (FTS) is a versatile electro-optical sensor for remote sensing, hyperspectral imaging, and laboratory chemical kinetics. Principal features include broad spectral coverage and high spectral resolution (Fellgate advantage) and high throughput (Jacquinot advantage). Due to its versatility, across various requirements, e.g. (resolution, bandwidth and aperture) sensor architecture contains an N-dimensional parametric trade matrix that needs to be readily assessed. Specifically considered are the logical steps utilized to flow down primary (customer) requirements and specifications to secondary (derived) requirements. Configurational aspects, generic trades, and parametric selections are emphasized for non-imagers as well as for imaging FTS. With an appropriately designed robust sensor, the noise equivalent spectral radiance or NE(Delta) N performance will be largely dictated by the scene and the instrument background flux. The performance will not be dictated by noise terms associated with interferogram encoding and signal handling. The mathematical formalism of interferometric error source types and photon limited design expressions are presented. The composition of these expressions are examined from the points of view of optical band limiting and some useful trade rules parametrically relating scan time and S/N to spectral resolution. For a well designed and executed interferometer, typical performance data are presented in terms of modulation index, calibrated radiometric atmospheric spectral signatures, and atmospheric spectral signatures for two spectral resolutions.