Sixty percent of all proteins are located inside cells. Many of these proteins are involved in pathways that regulate a variety of critical cancer cell survival and immunomodulatory processes. However, conventional macromolecular therapies targeting these intracellular pathways in cancer face several transport barriers including, tumor selective accumulation, dispersion, cell internalization and endosomal release. Therefore, an effective delivery vehicle is needed to circumvent the transport limitations associated with macromolecular therapies. Salmonella is an ideal intracellular macromolecular delivery vehicle for cancer therapy/immunotherapy. Non-pathogenic versions of the bacteria colonize and grow in tumors at ratios greater than ten thousand to one over any other organ. The bacteria are highly motile, disperse and efficiently invade non-phagocytic, epithelial cells. After cell invasion, Salmonella activate a unique set of genes selectively inside cells to upregulate type three secretion system-two activity, which, enables intracellular survival. This combination of traits is unique to Salmonella, making genetically engineered versions of the bacteria ideal for intracellular therapeutic delivery selectively within tumor cells. This thesis had two purposes. The first was to determine the critical driving mechanisms governing intracellular therapeutic delivery in tumor cells and genetically engineer a delivery strain of Salmonella based on this information. The second was to demonstrate, for the very first time, that the engineered Salmonella could deliver protein antigen into tumor cells and refocus preexisting, vaccine induced, immune cells to target cancer. We hypothesized that controlled expression of the master motility regulator, flhDC, in Salmonella drives tumor colonization, bacterial dispersion, invasion and protein delivery selectively inside tumor cells. To test this hypothesis, we employed a range of genetic engineering techniques, cell-based infection assays, in vitro tumor-on-a-chip and in vivo infection/tumor models to elucidate the driving delivery mechanisms of engineered Salmonella. Controlled expression of flhDC in engineered Salmonella enabled high levels of intracellular protein delivery selectively inside tumor cells. Once the delivery strain was created and optimized, we hypothesized that the engineered Salmonella could deliver the model vaccine antigen, ovalbumin, into tumor cells and refocus preexisting, vaccine induced immune cells to attack cancer. To test this hypothesis, we employed cell-based, syngeneic, and transgenic mouse infection/tumor models. Delivering the model vaccine antigen, ovalbumin, into tumors with engineered Salmonella refocused vaccine associated immune cells, including ovalbumin specific, CD8 T cells to combat and cure cancer in a subset of mice. This result demonstrates that Salmonella engineered to deliver vaccine antigen into tumors could be employed as an effective, off-the-shelf cancer immunotherapy with broad applicability in previously vaccinated cancer patients.