I am a graduate research assistant for the NSF-funded project "Syntactically-Annotated Corpora for Endangered Languages in Areal Contact", a collaborative research effort between IU and U of A.Coordinate annotation efforts for 5 endangered languages spoken in Mexico and GuatemalaExplore active learning for UD annotationDevelop NLP pipelines for endangered languages and explore areal contact phenomena from a corpus-based perspective.

Advise on data and modeling strategy for text-to-speech (TTS) system formonolingual and code-switched Anishinaabemowin and English.Train models and prototype approaches to low-resource code-switching TTS.Develop and maintain API server for calling TTS models in production.

NLP/Machine Learning to identify and extract key information from noisy student- and teacher-uploaded educational resources.

Contributed to fundamental research related to education and assessment, using state-of-the-art machine learning and NLP techniques. Research areas included automated constructed response and essay scoring, building robust language models for speech recognition, native language identification, and language and dialect detection.Worked on development and maintenance of operational automated constructed-response and essay scoring engines.Co-authored and presented research publications, and assisted in submission reviews for SIGDial, BEA, and other CL/NLP conferences.Contributed to ETS-maintained open-source libraries SciKit-Learn Laboratory (https://github.com/EducationalTestingService/skll/) and RSMTool (https://github.com/EducationalTestingService/rsmtool).

Owned NLP testing and evaluation, including regression testing, human surveys, and research for further linguistic development.Built web application for machine-assisted Named Entity disambiguation, using Python Flask, SQLite, and Javascript.Managed and maintained MySQL database of annotated linguistic corpora.Developed and maintained multiple internal tools providing access to large amounts of data from disparate sources in a straightforward, user-friendly way.Presented summaries of analyses to Product Management, Sales, and Marketing teams.

Developed Linguistic analytics tools, including custom part-of-speech tagger and statistical natural language generation models for search queries, using Python. Applied computational linguistics to identifying, selecting, and generating optimal keywords for Search Engine Marketing. Scraped, munged, and mined data for training machine learning models. Used Google AdWords scripts (JavaScript) and api (Python) for analyzing and managing SEM keywords.