Introduction

My journey to finding the world of Human Language Technology (HLT) began as far back as trying to teach myself Japanese as a middle schooler, and beginning to learn HTML by customizing my MySpace page like many others around my age at that time. These types of childhood hobbies formed my lifelong passion for language and technology, ultimately steering me in the direction of the perfect blend of the two—computational linguistics. I gained a solid foundation in the field during my undergraduate studies in English Linguistics at Arizona State University (ASU), complemented by minors in both German and Japanese. Then, after studying topics related to coding and obtaining a certificate in programming through Maricopa County Community Colleges, I began my Master of Science in Human Language Technology at the University of Arizona.

The topics I delved into during the HLT program have been vast, covering a wide spectrum of Natural Language Processing (NLP) techniques and linguistic theories. I continued learning about the fundamentals of formal language and grammar theories I was introduced to in my undergraduate degree. Additionally, I delved into the statistical backbone of modern NLP techniques, including n-gram models, information retrieval, smoothing, Hidden Markov models, and advanced algorithms such as Viterbi. Speech technology was also covered as part of the program, including a couple of projects I worked on for speech recognition and text to speech in Dutch. Other types of projects I worked on included sentiment analysis, document classification, and working on many fundamental NLP techniques such as building tokenizers and part of speech taggers. Throughout the duration of the program, I continued to develop my coding skills in Python, and also touched on PERL and Prolog. You can read about one of the many projects I worked on as part of the HLT program on my Portfolio page.

Before starting my MS, I already started my career in the education industry. I’ve worked solely in positions where I’ve been involved in technology, and in environments where I can continuously learn. My first job in the Accountability & Research Department at the Arizona Department of Education allowed me to develop skills in data analysis, including practical applications using SQL and SAS. I generated official public state reports and frequently prepared graphs used in presentations to the state board. My language skills also allowed for me to work as an editor for the monthly newsletter, and to be a main point of contact for school administrators across the state (showcasing my communication skills).

After working with the AZ Department of Education for two years, I transitioned to EdPlus at ASU as a Business Analyst, where I’ve gained a vital understanding of software development processes. While continuing to learn and grow as a Business Analyst at EdPlus, I was also offered a unique opportunity to intern with EdPlus at ASU’s Artificial Intelligence Product Department (AIPD) team. EdPlus’s AIPD is a small unit at the university that works with cutting edge technologies, and develops various applications and tools to enhance student journeys from start to finish. During this internship I gained a firm grasp on the concepts and applications of working with OpenAI’s Large Language Models (LLM) and Generative AI, including setting up Retrieval-Augmented Generation (RAG), experimenting with prompt engineering, creating chatbots, etc.

The internship also allowed me to develop my programming skills in a practical way, where I often had to learn how to accomplish small tasks as part of the bigger project as I moved along. A challenge I faced was having no experience with LLMs or OpenAI in particular at the start of the internship, but after getting to experiment, research, and use the openai Python library to produce valuable output, I now feel extremely confident with it. I found using OpenAI’s tools in Python can accomplish many NLP related tasks that I learned about in my program: tokenization, information retrieval, question answering, information extraction and summarization, creating vector embeddings for text, and more. Where the LLM sometimes fell short, I was able to leverage libraries like nltk that I already had exposure to and experience with through my HLT program. You can read about each project I worked on during the duration of my internship on my Portfolio page.

Now that I am about to complete my Master of Science, I can reflect on this journey that has been both challenging and rewarding. This path has not only honed my technical skills and deepened my understanding of computational linguistics, but has also inspired me to apply this understanding to real-world challenges. I want to use this technology to break down barriers for people so our lives can be less complicated, and build bridges for different communities. With a solid foundation in both theory and practice, my goal is to contribute to the development of technologies that enhance communication and understanding across linguistic and cultural divides, bringing the world closer together. As technology continues to evolve, I am excited about the endless possibilities to innovate and make a difference in this field.