Project EmpowerMD is an incubation team at Microsoft Healthcare. Our mission is to empower physicians by building a voice-enabled intelligent assistant, leveraging the best in speech-to-text, NLP, and other ML technologies. We are a fast-moving multi-disciplinary team that is deeply engaged in the healthcare space with ample opportunity to learn directly from clinicians and deliver broad impact.
WHAT WILL YOU BRING? You will be responsible to execute on the team's Natural Language Understanding (NLU) efforts as we aim to extract intelligence from human-to-human clinical dialogues in order to generate a schematized medical note. You will go deep on a range of tasks including data annotation, language processing, language generation, and dialogue management/ranking. You will help establish and implement data pipelines. You will have ample opportunity for cross group collaboration, technical product definition and applied research. You enjoy being part of a collaborative, multidisciplinary
REQUIRED QUALIFICATIONS: BS, MS or PhD in Computer Science, Statistics, Mathematics or related field 3+ years working on production ML data pipelines 2+ years of one or more of: named entity recognition, topic classification, knowledge graphs, dialog state modeling Fluency with recent NLP and ML advancements, e.g., word vectors, attention, LSTMs, etc. Strong coding skills in one or more of Python, C/C++, Java, Scala, C# PREFERRED QUALIFICATIONS: 1+ years with one or more of: NLTK, Stanford CoreNLP, spaCy, etc. 3+ years with one or more of: PyTorch, TensorFlow, Keras, CNTK, etc. 3+ years with one