Transition Edge Sensors (TESs) are ultra-sensitive superconducting detectors having a wide range of applications, from quantum cryptography to x-ray spectroscopy. Models of TESs often employ a circuit theory approach, taking TES parameters such as the temperature and current dependent resistance as inputs, but there are few microscopic models which allow prediction of these parameters. In this thesis, I develop a microscopic model to describe TESs based on the Usadel equations, allowing predictions of resistance surfaces, critical currents and the small signal electrothermal parameters. To test my model, I calculate I(V) curves for device designs used by different research groups and compare them to experimental measurements. Using my model, I design a TES wafer for a systematic study of the effects of bilayer size and patterning on performance. To investigate the effects of magnetic field, I also design and characterise a new cryogenic magnetic field test facility to apply fields of variable direction and magnitude. I present the results of this study, showing that bilayers of around 10-20 square micrometres present a good balance between low field susceptibility and predictable behaviour. I then investigate the use of a backing plate as an on-chip shield, to reduce the field susceptibility of the TESs by attenuating the perpendicular field component. I present the models used in this design process to ensure a good shielding factor as well as a high reflection coefficient, and repeat key measurements from my systematic study to show the reduction in field achieved using this prototype on-chip shield. TES readout typically uses Superconducting Quantum Interference Detectors (SQUIDs), placing limitations on the readout frequency. I investigate the possibility of using a High Electron Mobility Transistor (HEMT) amplifier in parallel with an inductor for readout, increasing the bandwidth and allowing higher frequency operation, and discuss how TESs might be optimised for this new readout scheme. Through this thesis I have explored a range of microscopic physics and discussed how this understanding can be used to enhance the performance of transition edge sensors.