Yosef Gavriel Tirat-Gefen., PhD
Rockville, MD ***53
571-***-**** (cell) ad8szr@r.postjobfree.com
Summary
Expert research and development professional in the areas of data science, artificial intelligence, operations research, supervised and unsupervised machine learning techniques, deep learning, reinforcement learning, data fusion, natural language processing (NLP), multi-objective optimization, requirements engineering, business and competitive intelligence, big data technologies, visualization techniques, forecast models, Bayesian modeling and techniques, stochastic modeling and optimization, systems engineering, quantitative risk modeling, scientific computing, parallel and distributed processing, mathematical modeling in systema biology, bioinformatics, biomedical signal processing Skills Summary DataRobot, ArcGIS, Scikit-learn, Snorkel, Numpy, Panda, Tensor Flow Software: PyTorch, Spark, Spacy NLP library, Alteryx, Tableau, Hadoop, AWS, Azure Machine Learning, Databricks, C/C++ SQL, Prolog, Lisp, Matlab, Scala, Docker, Cloudera, H2O, R, Python, SAS, Java, UML, ILOG, CPLEX, GUROBI, LP-SOLVE, GPU, Fortran, Oracle, Drools, JBoss, Jenkins, JIRA, GitHub.
Education and Training
• PhD, Computer Engineering, University of Southern California (USC), Los Angeles, CA.
• MS Applied Physics, George Mason University, Fairfax, VA.
• MS/BS (cum-laude) Electrical Engineering, Federal University of Rio de Janeiro.
• BS Applied Math, Computer Sciences and Information Systems, Rio de Janeiro State University.
• Native proficiency in Portuguese. Advanced knowledge of Spanish and French. Familiarity with Italian, German, Russian, Hebrew, and Japanese.
Relevant Experience US Citizen
MaxWave Research LLC., Northern Virginia and Rockville, MD Since February 2022 Senior Data Scientist
• Investigated alternative architectures for BERT and large language models (LLM) based recommendation systems for intelligence applications in a DoD agency.
• Developed mathematical models for survival analysis and reliability of smart devices for a power utility company.
• Designed a data science, machine learning and large language models (LLM) based software architecture to support NLP in legal AI applications.
• Deployed deep learning, reinforcement learning and generative AI techniques to score millions of border crossings for a DHS agency.
• Developed code in Python based in generative adversarial networks (GANs) for synthetic generation of tabular data for very imbalanced datasets.
• Deployed data science solutions in a graph database (Neo4J) including graph neural networks and community detection algorithms to identify network of rogue individuals for a border security application.
• Developed ensemble methods based in boosted trees methods, deep learning residual convolutional neural networks and deep learning transformer methods.
• Wrote reinforcement learning code to handle scoring and workload management of border crossings.
• Developed natural language generation (NLG) Python code to provide textual explanation of machine learning findings.
Defense Threat Reduction Agency (DTRA), Northern Virginia, VA November 2020 – April 2022 Research and Development Directorate / SAIC / ICESS /SPA Senior Data Scientist - Assistance and Advisory Services (A&AS)
• Advised senior decision makers on the development of intelligence analysis tools and the utility of data sets for natural language processing (NLP) in the Weapons of Mass Destruction (WMD) domain.
• Correlated publicly available information (PAI) with other data sets for natural language processing.
• Worked with data scientists and software developers to enhance data analysis methodologies.
• Operated and tested advanced prototype intelligence analysis tools.
• Evaluated the Snorkel.ai tool for computer aided labeling of large datasets in NLP applications.
• Investigated the use of graph neural nets to accelerate search in large graph databases.
• Developed AI/ML prototypes for natural language processing in intelligence applications.
• Performed terabyte searches in MongoDB to support intelligence analysts.
• Evaluated the use of Neo4J graphs to process network analysis for intelligence applications. Northrop Grumman Corporation (NGC), McLean, VA December 2019 – October 2020 US Department of State contract
MaxWave Research LLC / CompQSoft Inc.
Senior Data Scientist
• Designed and validated a machine learning pipeline using the Azure ML Designer visual tool for predicting the COVID pandemic growth.
• Designed, documented, and validated a prototype deploying ArcGIS Pro for geoprocessing of visa applications data from the US Department of State.
• Researched and wrote a research paper with recommendations for automatic monitoring, scheduling and allocation of cloud resources under cost constraints when deploying machine learning solutions in Azure, Azure Databricks and Azure Kubernete services (AKS).
• Reviewed and wrote research papers on state-of-the-art technologies and solutions in feature engineering, feature stores, automated machine learning (AutoML), feature engineering for AutoML.
• Developed computer vision and natural language processing (NLP) code to extract data from consular documents using the OpenCV and Tesseract OCR libraries in an Azure Databricks cluster. Reveal Global Consulting, Fulton, MD March 2018 – November 2019 Chief Data Scientist
• Secured multi-million dollars contract with the Center for Medicare and Medicaid Services (CMS) by prototyping a software tool demonstrating data science, statistical modeling, machine learning techniques and predictive analytics model development applied to detection of fraud, waste and abuse.
• Designed, validated and deployed data science code in the AWS-FedRAMP – US Federal Government secure cloud.
• Designed machine learning algorithms and code for data fusion of intelligence data for a DoD intelligence agency using natural language processing and large-scale graph algorithms.
• Investigated the use of network motifs algorithms in very large graphs and other machine learning approaches in intrusion detection and cybersecurity for US CYBERCOM applications in space, naval warfare and naval information systems.
• Acted as product owner in several Agile data science projects, providing detailed guidance including statistical quality assurance and dashboard design.
• Led a team of software developers, data scientists and business analysts. MaxWave Research LLC., Rockville, MD October 2016 – October 2020 Senior Data Scientist
• Developed cloud software, machine learning algorithms based on econometric FRED time series.
• Designed Tableau dashboards, and mathematical models in ethics compliance and human resources risk management, merging multiple data sets of both corporate and publicly available information (PAI). DeepMile Networks LLC., Arlington, VA August 2015- September 2016 Senior Data Scientist
• Led design and development of software for data science apps for health and sports league markets.
• Developed mathematical optimization methods using Gurobi/LP-Solve for sports league management.
• Wrote code in Java, C++, Python, SQL for deployment at a commercial cloud (Amazon AWS).
• Managed offshore software developers.
National Aeronautics and Space Administration (NASA) Metispace Technologies Inc., Rockville, MD June 2014- February 2015 Principal Software Engineer
• Developed mathematical optimization software in data science, operations research, scheduling, and allocation problems for aerospace applications.
• Deployed multi-objective optimization techniques, rule and knowledge engines, and rule based systems
(Drools) in aerospace problems.