Bernard Mcshea Email & Phone Number

Senior leadership, with proven strengths in analytical and novel problem solving, combined with experience and a deep understanding of cyber and machine learning. Critical skills needed to identify root causes of challenges, address key fundamental problems, and harness relationships between departments to actively manage programs and teams to discover efficient and innovative solutions to complex problems for effective identification of opportunities.Areas of Expertise:-- Harnessed diverse complementary background and granular knowledge of physical properties of hardware to distill complex challenges, address fundamental problems to find novel approaches to previously unsolved challenges such as data ergonomics, ransomware detection, and causal knowledge extraction.-- Reasoned Cyber Visualization developed predicate logic to generate tiered visualizations of cyber defense coverage and interactions of defenses addressing security operation center’s data ergonomics problem ensuring clear, concise and timely communication of cyber security risks to wide range of audiences.Key Achievements :-- Developed algorithms to construct neighborhood watch for electronic components connected via local bus, by retrofitting components to function as forensic sensors and detect malware. Late-Stage Cyber Attack Recovery seedling achieved 99% detection of ransomware and 92% detection of botnet denial of service attacks while only adding 6% additional computational overhead.-- Developed cooperative causal knowledge extraction via learning from diverse perspectives. Distributed Exploration with Aggregated Learning seedling, facilitating discovery of unknown confounding factors via semantic aggregation of knowledge and divers game play, enabling detailed reasoning.-- Formed and led BAE’s DARPA Cooperative Secure Learning (CSL) research team enabling learning on unbalanced NetFlow data while ensuring minimum impact on model’s final accuracy. Difference in accuracy scores for models trained on synthetic NetFlow data and real data and tested on real data ranged from -0.07% to +0.09%. -- Perceived issues and implemented discoveries to accurately label data for DARPA Cyber-Hunting At Scale (CHASE) applied to BAE Systems ESAFE system, achieving over 95% accuracy.Education:Graduate Certificate in Data Science, Harvard Extension School, Cambridge, Massachusetts M.S. in Computer Science, Department of Computer Science at Purdue University, West Lafayette, Indiana B.S. in Computer Science and Mathematics, Beloit College, Beloit Wisconsin Email: bernardmcshea@gmail.com