SUMMARY

• Principal AI architect and engineer with over twenty years of experience.
• Business acumen in a host of diverse industries, including high technology, banking, securities, insurance, retail, transportation, media, outsourced services, and healthcare.
• Expert in core AI related software architecture and engineering disciplines, including NLP, deep learning, and the simulation of human like reasoning.
• Track record of creating highly successful software products and services.
• Extremely broad range of skills that have consistently created new revenue streams, reduced costs, increased profitability, and provided clients with a competitive edge in the markets they serve.

TECHNICAL SKILLS

• Artificial Intelligence, SOA, SOAP, XML, Spark, Scala, Ignite, DATA GRID, AWS, Azure, Cloud, Hadoop, Flume, Hive, HDFS, HBase, MongoDB
• Cassandra, Big Data, Kafka, data pipeline, Natural Language Processing and Understanding, Definite Clause Grammars
• BRMS, business rule engines, Prolog, speech recognition, Named Entity Recognition (NER), speech synthesis, Robotic Process Automation (RPA)
• Current and Future State Architecture, Physical and Logical Architecture, Linux Intelligent Virtual Assistants, enterprise architecture
• Java, Cognitive Computing, Python, learn - imbalanced, scikit-learn, Pandas, NumPy, Deep Learning, H2O, Tensor Flow, Keras
• Parts of Speech (POS) tagger, predictive analytics, OOAD.

PROFESSIONAL EXPERIENCE

Confidential

AI Architect

Responsibilities:

• Developed the logical and physical architectures for an intelligent virtual assistant (IVA) that would work right alongside a clinical review nurse. The purpose was to augment their capabilities, as well as increase overall productivity. The IVA would electronically read through EMR notes, such as ER, history and physical, discharge, and specialist consults, in order to determine if clinical necessity guidelines for hospital admission had been met. The system incorporated a clinical Named Entity Recognizer (CNER), a grammar and vocabulary tailored specifically for medicine, as well as a document planner and natural language generation subsystem that were responsible for the actual writing clinical appeal letters.
• Conceived an enterprise level reference architecture for artificial intelligence, that seamlessly integrated machine learning (ML), robotic process automation (RPA), natural language understanding (NLU), and operational decision support (ODM)
• The architecture incorporated an abstraction layer as well as a cognitive computing engine that combined mathematical approaches to predicting outcomes along with human like heuristics.
• Developed the logical architecture in support of the overall vision of the platform, identifying all of the required components and subsystems, including their individual responsibilities.
• Drilled down into the physical architecture for the distributed real-time computing subsystem, identifying a combination of open source initiatives that would provide the ability to virtualize CPU, memory, and disk storage resources, across an entire data center, in addition to geometrically scaling out the execution of machine learning models using a peer to peer computational GRID
• Worked with IS business partners to identify, document, and rank potential use cases that could serve as the business requirements for a highly visible proof of concept
• Designed and built a proof of concept that combined both optical character recognition, along with deep learning based text classification, to automatically categorize written correspondence received from payors. In order for it to be moved into a work queue. Utilized Tesseract and its core LSTM based OCR engine to extract the textual content of scanned documents being stored as .TIFF images. Built the deep learning models using Tensorflow, Keras, and BERT (via ktrain)
• Developed prototype code, to demonstrate and showcase the capabilities of image based deep learning in the healthcare space. The model was capable of classifying greyscale images of chest x-rays with ‘near human’ accuracy

Artificial Intelligence Architect

Confidential

Responsibilities:

• Researched and assessed leading approaches to building deep learning models that would analyze medical images in order to make a diagnosis. Using sample x-ray and blood smear datasets from the National Library of Medicine and National Institute of Health, adapted open source implementations in order to maximize the recall rate for the classification of pneumonia and malaria images. Recall rate for the classification of pneumonia x-rays approached 96%.
• Developed a detailed business case and reference architecture, for the use of machine learning, natural language understanding, and human like heuristics to improve member population health. The goal was to delay the onset, as well as reduce the mortality and morbidity rates normally associated with strokes and diabetes. This would be accomplished through the identification of high risk groups and the subsequent creation of intervention strategies targeted towards affecting (delaying if possible) the medical outcome.
• Developed a Natural Language Processing/Understanding (NLP/NLU) front end for an intelligent virtual assistant that could assess the risk of stroke. Utilized Prolog’s Definite Clause Grammar (DCG) representation to evaluate the syntactic structure of sentences, ensuring they conformed to a standard language definition. Extended the DCG grammar rules to handle the morphological structure of words including verb conjugation, plurality of nouns, and subject verb agreement. Through the use of lambda calculus expressions, incrementally built a knowledge representation of the sentences during syntactic parsing, This code created, and stored, the semantic meaning of the sentence. Incorporated logic based constraints to ensure the pragmatic requirements were being met, based upon the sentence context.
• Used scikit-learn, learn-imbalanced, and pandas, to build a highly accurate machine learning model that correctly identified 5 out 6 plan members who were likely to experience a stroke at some time in the future.
• Constructed data pipeline used to manipulate raw training and test data. Dropped rows having large numbers of missing values, imputed mean, and performed one hot encoding of nominal values.
• As the original dataset was imbalanced (98%, 2%), used random, over, and under sampling, as well as the SMOTE algorithm, to ensure that there were an equal number of positive and negative target values within the training dataset.
• Built predictive models using logistical regression, random forest, and support vector machine algorithms. Compared model performance using a confusion matrix and the calculation of precision, recall and F1 scores.
• Reviewed potential AI use cases, across different lines of business and operating groups within a health insurance environment, including compliance, claims processing, and member health. The goal was to identify high value opportunities for cognitive automation, that would readily lend themselves to a proof of concept whose chances of success were very high.
• Built a “No Cost” Robotic Process Automation (RPA) environment from scratch, using common off the shelf infrastructure components. This included the JADE mobile agent platform, which provided a highly scalable, container based system for executing, managing, and monitoring the entire lifecycle robotic software processes. Also incorporated the jBPM business process manager, in support of externalization of both business logic and robotic workflows.
• Developed a lightweight RPA framework (set of API’s), that “wrapped” the above mentioned open source infrastructure components. This simplified the coding effort required, and reduced the complexity of writing, any type of RPA application.
• Using the “homegrown” RPA environment built a proof of concept to automate some of the laborious manual processes that auditors perform within the corporate compliance group.