Biosensors capable of tracking the body's molecular state have tremendous potential to improve lives by making healthcare more accessible, personalized, and preventative. At the same time, developing biosensors that are both compatible with point-of-care or continuous monitoring and can accurately track a wide range of clinically important biomarkers poses a tremendous technical challenge. These sensors must operate without the sample processing employed by conventional laboratory diagnostic assays and must measure targets rapidly with molecular sensitivity and specificity within complex biofluid samples. These demands may be met by biosensors based on "molecular switches" -- engineered receptors that bind molecular targets with high specificity and undergo a binding-induced conformational change that produces measurable optical or electronic signals. However, to date these molecular switch biosensors have remained limited to sensing only a handful of molecules. It remains an outstanding challenge to develop generalizable strategies to methods for engineering molecular switches to sense the complete range of health biomarkers. In this dissertation, I begin by considering how we can develop sensors that are compatible with real world, day-to-day use. I analyze the diagnostic value of biosensing in dermal interstitial fluid, a biofluid with emerging interest for use with minimally invasive and continuous wearable sensors. The remainder of the dissertation focuses on three projects where I expand the generalizability molecular switches, enabling more rapid development of biosensors for new biomarker targets. In the first project, I introduce an approach for engineering off-the-shelf antibodies into antibody-based molecular switches that achieve continuous optical biosensing by augmenting them with a rationally designed DNA-linked competitor molecule. In the second project, I develop a universal protein-based competitor molecule that provides an even more universal method for our antibody-based molecular switch design to be applied to sensing new targets. In the third project, I develop a rational design approach for engineering DNA-based molecular switches that have pH-dependent properties which can be tailored to improve biosensor performance, and for potential drug delivery applications. Taken together, these advances in molecular switch design offer a toolkit of generalizable strategies that can be used to develop next-generation biosensors which sense a wide range of biomarkers for addressing a wide range of clinical needs.