Summary
Peter Zukerman is a Natural Language Processing scientist with eight years of experience building scalable text and speech systems, currently driving ASR and NER improvements at DCS Corp. He combines computational linguistics training from the University of Washington and a CS/linguistics BS from UIUC to deploy deep learning and transfer-learning solutions that measurably cut ASR word error rates and boost entity precision on low-resource datasets. His research background spans low-resource MT, data augmentation, and a BlackBox NLP paper on verb alternation embeddings, plus interdisciplinary work on how anthropomorphized language affects public trust in AI. Comfortable across the stack, he has production experience with AWS-backed APIs and mobile deployments, and earlier work enlarged under-resourced speech corpora by an order of magnitude. Multilingual in Russian and Japanese (conversational Korean and Spanish), he writes about NLP and linguistics on his blog, bringing both engineering rigor and language-first intuition to applied AI problems.
8 years of coding experience
4 years of employment as a software developer
Master's degree Computational Linguistics, Master's degree Computational Linguistics at University of Washington
Bachelor of Science - BS Computer Science and Linguistics, Bachelor of Science - BS Computer Science and Linguistics at University of Illinois Urbana-Champaign
Russian, English, Japanese, Korean, Spanish