Summary
Alexandra Teste is a Principal Data Scientist with 11 years of experience building production ML systems that bridge research-grade models and product impact, currently at LexisNexis Legal in Bellevue. She has led end-to-end efforts from data labeling and preprocessing to fine-tuning transformers and deploying multilingual and multimodal models in production across companies like Dataminr, OfferUp, and Hirefly. Her background in space plasma physics and academia gives her a rare aptitude for breaking complex problems into tractable analytic components and communicating technical trade-offs to nontechnical stakeholders. She combines hands-on engineering (Airflow, Spark, FastText, sentence-transformers) with team leadership—setting up ML workflows, mentoring, and running cross-functional rituals to align product and engineering. Notably, she’s applied transformer fine-tuning not only for translation and sentiment but also for practical hiring-matching systems and embedding-based title recommendation. Her work emphasizes rigorous data hygiene, reproducible pipelines, and extracting measurable business value from advanced models.
11 years of coding experience
13 years of employment as a software developer
Classes Préparatoires aux Grandes Ecoles Physics-Chemistry, Classes Préparatoires aux Grandes Ecoles Physics-Chemistry at Université Paul Sabatier Toulouse III
Master's degree Astrophysics, Master's degree Astrophysics at Paris-Sud University (Paris XI)
Ph.D. Space Plasma Physics, Ph.D. Space Plasma Physics at Pierre and Marie Curie University
Master's degree Geophysics, Master's degree Geophysics at Ecole de Physique du Globe de Strasbourg (EOST)
French