My work involves the notion of encoding arbitrary information into vectors that can be used as features for downstream systems, or on their own, for inference and learning. My goal is to allow different modalities of data to exist, and combine together, in a single space of vectors, that is both geometrically and semantically sensible. This involves using Hyperdimensional Binary Vectors and Hyperdimensional Computing (first proposed by Kanerva) to achieve representations of any data as long, binary vectors that have very useful mathematical properties. With some clever minimization and structure, many forms of information can be represented as always constant-sized, meaningfuly constructed binary vectors. My main interest is to used this paradigm to allow language and perception to coexist in the same space, meaning I apply this technique to both linguistics tasks and computer vision tasks. I also work with embeddings in general, such as word embeddings from word2vec and GloVE, researching how to better utilize them and contextualize vectors for use in a specific context. I work with Professor Yiannis Aloimonos in the Computer Vision Lab at the Department of Computer Science in University of Maryland, College Park. I also work with Dr. Douglas Summers-stay at the Army Research Lab (ARL) in Adelphi on using embeddings to enable reasoning in both visual and linguistic domains.