I specialize in computational linguistics and natural language processing (NLP), focusing on enhancing models for Portuguese language understanding. My experience includes fine-tuning machine learning models, annotating and analyzing linguistic data, and developing automated tools to support language analysis. I’m dedicated to using data-driven insights to explore language complexity, improve natural language understanding, and support linguistic research through a combination of quantitative analysis, corpus linguistics, and programming.
📖 Education
- 2025 - 2026, Master’s degree in Education candidate in the Learning Design, Innovation, and Technology program. Harvard Univeristy, Cambridge-MA. Advisor: Karen Brennan.
- 2016 - 2018, Master’s Degree in Applied Linguistics. Universidade Federal de Minas Gerais, Belo Horizonte-MG Thesis: Lexical Bundles in Sections of Applied Linguistics Research Articles: Advisor: Prof. Deise Dutra.
- 2006 - 2010, Bachelor’s Degree in English. Universidade Federal de Minas Gerais, Belo Horizonte-MG
📄 Publications
-
Erjavec, T. et al. ParlaMint: Comparable Corpora of European Parliamentary Data. CLARIN, 2021.
-
Dutra, D., Queiroz, J., Macedo, L., Costa, D., Mattos, E. Adjectives as Nominal Pre-modifiers in Chemistry and Applied Linguistics in Research Articles - a chapter in Advances in Corpus-Based Research on Academic Writing: Effects of discipline, register, and writer expertise: ed. by Ute Römer, Viviana Cortes, Eric Friginal. John Benjamins Publishing Company, 2020.
-
Macedo, L., Costa, D. Lexical Bundles Conveying Modality in Argumentative e Essays of EAP Students: a corpus-based study. ANAIS DO X CONGRESSO INTERNACIONAL. ABRALIN, pp. 2068-2076, 2017.
📖 Work Experience
-
10/21 - 02/23, NLU developer for Cerence: Development and testing of grammars for NLU (pt-BR and pt-PT); text manipulation and data processing; localization of data; accuracy and coverage improvement; prepare and release models biweekly.
-
10/18 - 08/20, Associate Linguist for Google via Lionbridge: Linguistic expertise in developing NLP tools for Brazilian Portuguese. My position mainly involved developing grammars; creating templates for Natural Language Understanding; evaluating current system outputs; annotating and reviewing linguistic data; labelling text for disambiguation, expansion, and text normalization; research, development, and experimental projects.
-
08/18 - 10/18, Technical writer for LexisNexis Risk Solutions: Writing, translation, editing, and proofreading of IT and Software manuals and procedures, En-Pt and Pt-En.
🍎 Teaching Experience
- 11/23, Workshop Corpus pre-processing and cleaning using CLI(command line interface) for Universidade Federal de Minas Gerais, under the supervision of Prof. Deise Dutra.
- 08/12 - 08/18, English and Portuguese (L2) teacher for Rizvi Intl Schools Classes tailored for business students at multinational companies.
- 01/16 - 08/16, Instructor of English for Academic Purposes for Universidade Federal de Minas Gerais: graduate and undergrad students from UFMG. IFA project, under the supervision of Prof. Deise Dutra.
💻 Current Projects
-
working on the annotation of a legal text in Brazilian Portuguese (LEX-BR-Ius) with Professor Lucia Ferrari from UFMG, using UDPipe - our aim is to investigate diachronic language change since 1943 (the creation of the document) until nowadays.
-
volunteer researcher in the analysis of academic writing in the humanities with Professor Deise Dutra and Professor Tony Berber Sardinha.