Durga Prasad Sanugula Email & Phone Number

Los Angeles , California , Us

Seattle, Wa, Us

AWS Infrastructure Supply Chain and Automation

Seattle, Wa, Us

Infrastructure Supply Chain and Automation

Minneapolis And St. Paul, Minnesota, Us

•Managed a huge public health client system with millions of records using Omaha System SQL database. •Built Machine Learning models to classify clients suffering from pain by leveraging data from Omaha System.•Languages: Python, R, SQL

Worked on Bixby Project (Samsung's virtual assistant) for categorization of Android applications and their features from the UI syntactic navigation flows•Classified Android applications UI navigation flows to associate them with various entities and actions.•Performed feature engineering and clustering to extract the meaningful insights from raw app sdk data and automate data labelling•Developed and tuned various Machine Learning models like Support Vector Machines (SVM), Gradient Boosting (GBM), Random Forests(RF) and Multilayer Perceptron (MLP).•Designed and developed various Deep Learning models like Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs) with LSTM units and tested the models performance with various feature engineered attributes generated using autoencoders and embedding techniques. •Techniques/Languages used: Python (Pandas, NumPy, scikit-learn, TensorFlow, Keras), LSTM, CNN, SVM, GBM, RF, MLP

•Worked as a developer in S Voice and Bixby Projects on server side which is an essential element for intent determination and dialogue management of Samsung's virtual assistant.•Worked on dialogue management by implementing Natural Language Generation (NLG) flow using C++ for Samsung Connect agent in Bixby Project. Developed two different communications in Natural Language Understanding (NLU) server, one from NLU server to IoT server and the other from NLU server to Bixby client. It helped in achieving reduction of server response latency. •Trained Recurrent and Convolutional Neural Network models for Intent Determination in Bixby Project.•Designed and implemented NLG and contextual user interaction flow in C++ for Calendar, SmartThings domains and added support for the follow-up contexts. Responsible for improving the Calendar domain accuracy by 8%.•Collaborated with other teams (Client and Platform) to integrate and develop a voice based common platform for some of Samsung Digital Appliances. Digital Appliances can be made work by giving voice commands through Samsung Gear•Technologies/Languages used: C++, Python, Java, NLP, NLG, Regression

•Worked on implementation of various morphological processes like word segmentation, word decomposition, stemming, phonetic transcription, POS Tagging which are part of Natural Language Processing.•Worked on CYK and Charniak syntax parsing methods to identify the belongingness of a sentence to a particular language grammar.•Technologies/Languages used: C++, Python, Java, Morphology, Parsing, statistical Natural Language Understanding (NLU)

Master'S Degree, Computer Science

Bachelor’S Degree, Cse