Self-taught Senior Software Engineer with a Liberal Arts background and a wide breadth of experience in cloud infrastructure and data storage technologies.
Abiding interests include music, literature, film, linguistics and mathematics.
I believe we can come to a deeper understanding of scientific data and mathematical phenomena by experiencing them in novel sensory configurations outside of a typical data analytics dashboard.
Data Mapping conventions resulting in audio, haptic feedback, or other sensory input to which our minds are highly attuned, like faces, can enhance our intuitive understanding of concepts that are difficult to express with language and inform further more objective study.
AI was used for research, but not for code generation in the making of this site
My understanding of LLMs is informed by modern practice, as well as the fundamental questions like symbol grounding first proposed by linguists and philosophers like Pierce and Saussure over a century ago. I believe responsible LLM usage requires accounting for its effect on social structures and human collaboration in order to maintain an engineering culture that fosters creativity and excellence.
From an information theoretic perspective, the processes of predicting the next most likely token and data compression are indistinguishable. Therefore, I find it helpful to think of Ai/LLMs as highly compressed databases, which are ultimately built from human content curated by usefulness. While they can supply a nearly infinite breadth of knowledge, discernment must always be supplied by the operator.