Hire Logen F. - Forward Deployed Engineer

Lead development of representations of persons, extracting, normalizing and enriching data for 35+ selectors and 60+ datasets Lead development and pipeline management of ER model running on 15+ billion records across 30+ enriched datasets to disambiguate people and organizations Leverage deterministic matching, Bayesian probabilistic matching, connected components, transitive closure, and statistical scoring Use ML Flow and analytics to develop blocking rules to reduce candidate pairs for ER with significant coverage while optimizing computational load and processing time Support development of text analytics platform, enabling users to visually interact with text data to identify patterns, perform training data curation, and perform Natural Language Processing (NLP) tasks Investigate training data coverage and NER performance on 150+ entity classes Key contributor on suite of analytics incorporating AI/ML solutions to perform cognitive tasks across 10+ social media platforms Develop, manage and maintain multiple analytics and their pipeline, including relationship analysis, word cloud, and IP correlation Explore new capabilities using social media data such as image similarity, image to text and multi-modal large language models to improve discovery and investigation Manage team of data scientists, ML Ops engineers, data engineers, and developers distributed between Northern Virginia and Huntsville, Alabama Lead team during transition period between contracts, multiple client product owners, and maintain performance level and provided analytics with reductions personnel and significant development changes Responsible for presenting sprint status to section, and providing demos and overviews to leadership in Agile environment Conduct regular interviews of data scientists, cloud & ML Ops engineers, and data engineers.

Work on development team designing entity resolution model and platform as a data scientist. Assist in ingesting new datasets using Scala and Apache Spark and running ER within Databricks. Continually look for improvements to model to improve performance and quality of results. Developing a proof of concept, using Databricks, to design a refined ETL process within the DHS. Taking business rules for current ETL process written in plain English and converting them to Pyspark and SQL, to create over 100 tables with varying levels of complexity. Act as subject matter expert for Apache Spark and Databricks both within my project team, and the greater Booz Allen team.