Cancer is the second-leading cause of death worldwide. Over the past two decades, chimeric antigen receptor (CAR) T cell therapy has emerged as a promising alternative to traditional surgical, radiation and chemotherapy cancer treatments. Genetically engineered CAR T cells are designed to target and eradicate cancer cells in vivo. However, it remains difficult to identify a set of truly cancer-specific surface antigens to target—a critical requirement to prevent potentially fatal CAR T cell on-target off-tumor toxicity against other healthy tissues elsewhere in the body. I develop a variety of CARs and Receptors and assess their function using genetically encoded fluorescent protein-based biosensors to rapidly detect the pre-transcriptional molecular events leading to CAR-mediated T cell activation. I next propose the novel concept of using light to spatially and temporally limit CAR expression in T cells localized to the tumor site in order to limit on-target off-tumor toxicity in distant healthy tissues. After creating and evaluating a variety of light-sensitive protein-based optogenetic systems to control CAR expression, I uncover three limitations. First, even when kept in the dark, some light-sensitive engineered T cells prematurely express CAR. Second, engineered T cells stimulated with light only weakly upregulate CAR expression. Third, the amount of blue light exposure necessary to induce CAR expression is phototoxic to the T cells. To overcome these limitations, I create the first light-inducible optogenetic system capable of driving robust CAR expression in T cells only following stimulation with minimal, non-toxic amounts of blue light. To do so, I create and optimize a novel genetic AND-gate by integrating components of tamoxifen-inducible Cre recombinase systems with a blue light-inducible split Cre system driven by heterodimerization between the highly sensitive Magnet system protein domains, nMag and pMag. To prevent premature CAR expression, the cytosol-localizing mutant T2 estrogen receptor ligand binding domain (ERT2) is fused to the N-terminal half of the CreN-nMag fusion protein, thus physically separating it from its nuclear-localized binding partner NLS-pMag-CreC. Without tamoxifen to drive ERT2-CreN-nMag protein translocation into the nucleus, the high levels of spontaneous, premature Cre-loxP recombination native to the original photoactivatable split Cre system is significantly suppressed. Upon stimulation with both tamoxifen and blue light, T cells engineered with this novel optogenetic system undergo efficient Cre-loxP recombination to express CAR, with high sensitivity to low-intensity, short-duration blue light exposure. I demonstrate that the new tamoxifen- and photo-activatable split-Cre recombinase system, called TamPA-Cre, can be applied to strictly control localized CAR expression and subsequent T cell activation. The TamPA-Cre system has the potential to limit on-target off-tumor toxicity against distant healthy tissues in a way that was not previously possible.