• Contributed building ontology, customer intent and NER (Name Entity Recognizer) models using machine learning algorithm (Conditional Random Fields) for different user domains (e.g. Weather, Address Entry, Timer etc) which was at the core of Automatic Speech Recognition (ASR) engine [ Python, Bash] • Integrated customer intent labels obtained from trained models into databases• Conducted batch testing and optimization of trained models to improve classification accuracy to above 90% • Deployed fully trained and tested NER models onto server for testing of other domains