The free electron laser (FEL) is the brightest available source of x-rays, surpassing other options by more than ten orders of magnitude. The FEL's short ($\sim$femtosecond), high power ($\sim$gigawatt), coherent x-ray pulses are uniquely capable of probing ultrafast and ultrasmall atomic and molecular dynamics and structure, making them an invaluable research tool for biology, chemistry, material science, physics, medicine, and other fields. Unfortunately, all extant x-ray FELs rely on long rf linacs and undulators, with a footprint of kilometers and a cost on the order of a billion dollars. This severely limits the number of x-ray FELs, with the half dozen existing installations funded at the nation state level. These facilities are significantly oversubscribed, to the detriment of scientific and technological progress. Therefore, attempts to reduce the size and cost of FELs are an active area of research in an effort to increase access to these powerful research tools, with the goal of making x-ray FELs affordable to universities and companies.

One of the approaches being researched is the laser plasma accelerator (LPA). The LPA uses an ultra-high intensity laser to eject plasma electrons from a bubble region, producing longitudinal accelerating fields more than three orders of magnitude higher than what can be achieved in an rf linac. In principle, this could shrink the FEL accelerating section from the kilometer scale to a tabletop. To date though, despite continual progress and refinement over the last decade, LPA beam quality has not yet reached the level where it can be directly used as an FEL driver due to stringent constraints on the lasing dynamics.

The BELLA FEL experiment at Lawrence Berkeley National Lab intends to decompress the beam to skirt some of the beam quality requirements, by stretching the beam longitudinally and reducing local energy spread. This dissertation will discuss the design and implementation of two subsystems essential for the successful operation of this experiment. The first of these is a coherent transition radiation bunch length diagnostic, which is required to measure the length of the LPA bunches and extrapolate other details about the experiment's performance. The second is an electromagnetic chicane which performs the decompression of the electron beam. A final chapter explores the use of advanced undulators to enable the next generation of LPA driven FELs without decompression and discusses methods for realizing such undulators.