I am a data scientist, with a focus on human-computer interaction research, ML engineering, and natural language processing.I've been developing ML-powered systems since 2014. I started out as a software developer on R&D teams (MITRE, General Dynamics). During my Computer Science PhD at the University of Minnesota, I collaborated with multiple organizations (Amazon, CaringBridge, Wikimedia Foundation) to produce original computer science research, publishing open-source code and peer-reviewed publications. At ConcertAI, I deployed production ML systems in my role as a Senior Machine Learning Engineer on a data science team. At the Digital Harbor Foundation, I am deploying prototype machine learning systems based on large language models and conducting quantitative research.Skills:•Primary programming languages: Python, Java, R, JavaScript (I also have a soft spot for Clojure)•Frameworks and packages: PyTorch, Pandas, Matplotlib, scikit-learn, Hadoop MapReduce, Spark, SpaCy, React, Transformers, NLTK, gensim, networkx, statsmodels, NumPy, Flask•Machine learning methods: classification (logistic regression, SVMs, deep learning, XGBoost), regression (linear, Poisson), clustering, dimensionality reduction (e.g. SVD, PCA)•NLP (natural language processing) methods: embeddings, document representations and indices (TF-IDF), topic modeling, transformers (BERT/RoBERTa), Vowpal Wabbit•Qualitative methods: annotation codebook design, thematic content analysis, survey design, semi-structured interviews•Tools: Git, Jupyter, Vim, VS Code, LaTeX, SQLite, MySQL/MariaDB, MongoDB, Linux, PostgreSQL/Amazon Redshift, Docker•Domains: social media, online communities, annotation and crowdsourcing, social support, cancer care, Wikipedia & peer productionVisit my personal site for more about me and my research: https://levon003.github.io/