Next Century Corporation proposes the development of Muggsy, a low-shot deep learning detection prototype system that learns to recognize uncommon targets in remote imagery. Our Phase I research extends and leverages an image classification system of our own design called EvoDevo. EvoDevo evolves its own neural network architecture before training to meet the complexity of the data. Muggsy uses learned features from a standard DCNN trained on massive sets of social media photos. We replace the final layers of the DCNN with EvoDevo and train it on the low-shot exemplars and a large collection of unlabeled remote images. This will allow Muggsy to outperform other computer vision systems in low-shot target detection and adapt to different sensor modalities (e.g., EO, MSI, and SAR). In addition, in Phase II we will extend our image generation algorithm SIGHTT to create more realistic images to extend the low-shot exemplars. We will evolve an adversarial neural network to transform our typical synthetic image to become more like real images. In this manner, we will be able to extend the training data to offer better detection capabilities on targets with few examples.