Computational Linguist & NLP Software Engineer

Stefan Bartell

******.*******@*****.*** · 320-***-**** · www.linkedin.com/in/stefanbartell/

Computational & Experimental Linguist · Cognitive Scientist · Software Developer. 12 Years Python seeks careers: Computational Linguist, Language/NLP/Prompt Engineer, i.a. WORK & RESEARCH EXPERIENCE · skills · projects · references upon request 1. Snorkel AI: LLM/Chatbot Prompt Engineer, Linguistics/Cognitive Science Expert 01/2025-03/2025

· Large Language Model, Academic Teaching, Semantics, Syntax, Phonology, Artificial Language Learning

· Engineer prompt questions (10 answers) to stump ( 20% accuracy) & train frontier LLMs sold to clients e.g. OpenAI using exam-writing & human subjects research skills. Experiment to increase difficulty. Pioneer prompts reasoning about artificial languages, novel morphemes. Mostly independent 2. Weill Cornell Medicine: Computational Linguist, Postdoctoral Researcher 08/2022-02/2024

· Natural Language Processing, Context-Free Grammar, Syntax, Semantics, Named Entity Recognition, Statistics; Data Analysis, Annotation, & Visualization; Prodigy; Python: Pandas, spaCy, Matplotlib/Seaborn; Regular Expressions, Algorithm Development & Optimization, Protégé Ontologies

· Medical & computational linguistic research: Identify in text via algorithm with GFG dosage phrases

(dose, substance, … e.g. '200 mg aspirin'). Develop software to implement, test, & improve precision & recall of algorithm. Research how stance (positive/negative/neutral) on 2,4-dinitrophenol (DNP) predicts substance mentions. Quantify usage patterns & effects across substance mentions & doses extracted from social media & forums. Construct chemical ontologies. Engineer prompts to review literature & draft code. Control versions in GitHub. Write manuscripts & presentations. PI: Michael Chary (MD, PhD) 3. U. of Delaware: Grad Student Researcher, Language Acquisition & Brain Lab 08/2019-05/2020

· Data Science & Structures, REDCap, Cognitive Neuroscience, Psycholinguistics, Python: Pandas

· Develop software to integrate many database formats into one, Research Electronic Data Capture. Investigate role of hippocampus in statistical learning reviewing studies using e.g. functional magnetic resonance imaging & write research ideas. PI: Zhenghan Qi (MD, PhD) 4. U. Delaware: Dissertation Researcher, Linguistics & Cognitive Science 04/2017-04/2021

· Semantics, Syntax, Cognitive Science, Experimental Psycholinguistics & Design, Anaphora, Python: Pandas, Statistics with R, Data Analysis & Visualization, Amazon.com Mechanical Turk

· Design & execute experiments on sentences with 'donkey' anaphora that challenge traditional semantics with source of ambiguity & logic representation unclear. Subjects judged sentences true/false within scenarios, indicating meaning. Novel finding: effect of indefinite phrase determiner e.g. 'some/a/any squirrel' on interpretation corroborated by cross-linguistic comparison. Analyze how ambiguity is derived & meaning computed. Advisors: Satoshi Tomioka & Fred Adams

5. U. Delaware: Grad Student Researcher, Language & Cognition Lab 09/2014-03/2017

· Experimental Psycholinguistics & Design, Cognitive Science & Psychology, Statistical Learning, Syntax, Semantics, Data Analysis & Visualization, Python: Pandas, Statistics with R

· Design, implement, & write papers on artificial language learnability experiments on semantic systems based on evidential systems in languages worldwide. Create & present image (puppet) & video (human) events of 3 characters interacting using PsychoPy to teach adult subjects systems (1-3 novel morphemes). Direct research assistants to enact events. PI: Anna Papafragou 6. U. Delaware: Instructor & Teaching Assistant, Linguistics & Cognitive Science 08/2014-05/2021 Intro.: Cognitive Science (5 times), Communication Disorders in Children, Linguistics, Psychology; Language, Mind, & Society; Brain & Behavior, Measurement & Statistics (2 times), Sociolinguistics 7. University of Chicago: Research Assistant, Language Processing Lab 07/2012-09/2013

· Experimental Psycholinguistics, Data Analysis & Annotation, Statistics, Amazon Mechanical Turk

· Set up & run experiments on syntactic priming in ellipsis: MTurk & speaking/reading monitoring brain activity using event-related potentials. Create stimuli for experiments on semantics & on-line processing of adjectives. Write code in Python & R to transcribe & analyze data. PI: Ming Xiang LANGUAGE, COMPUTER, & OTHER SKILLS

· Native: English, Advanced: French, Intermediate: Japanese & Polish, Beginner: Spanish

· Prompt Engineering, Conversation Design, Languages: Haskell, Racket, Prolog, LaTeX, C, SQL, JavaScript, CSS, HTML; PsychoPy, Disjunctive Datalog, MATrix LABoratory, Praat, PyTorch

· Semantic Networks, Knowledge Graphs, Taxonomies, Language Processing, Philosophy of Language EDUCATION

1.Pre-medical, various sciences, Universities: St. Cloud State & Wisconsin-Madison. 01/2021-08/2022 2. Doctor of Philosophy, Linguistics, University of Delaware 08/2014-04/2021

· Courses: Computational Linguistics, Artificial Intelligence · GPA: 3.88 3. Bachelor of Arts, Linguistics: General Honors, University of Chicago 09/2009-06/2013 SELECTED POSTERS, PUBLICATIONS, & TALKS

1. SB & Chary, M. How does rhetorical stance affect content in online discussions about DNP use? European Association of Poisons Centres & Clinical Toxicologists, Palma de Mallorca, 2023 2. SB & Chary, M. Opinion influences content in social media discussions of DNP. ACMT, San Diego, 2023 3. Saratsli, D., SB, & Papafragou, A. Cross-linguistic frequency & learnability of semantics: Artificial language learning studies of evidentiality. Cognition, 197, 2020 4. SB. Effect of indefinite form on donkey anaphora interpretation. MACSIM, NYU, 2019 5. SB & Papafragou, A. Cross-linguistic variation & learnability of semantic systems. BUCLD, 2015

Contact this candidate