Charles Twardy is a Principal Data Scientist and affiliate professor with 12+ years applying Bayesian and probabilistic methods to AI/ML problems in cybersecurity, forecasting, and search-and-rescue. He has led multi-institution DARPA and IARPA programs—building scalable crowdsourced Bayesian prediction markets and real-time coordination across thousands of participants—and shepherded projects that spun into operational cyber and intelligence tools. His work blends formal philosophy-of-science rigor with hands-on machine learning and NLP, producing classifiers for cardiovascular risk, counter-IED attribution, and lost-person behavior, plus published work on argument mapping and causal representation. Known for expertise in hierarchical Bayesian models, causal inference, and structured analytic techniques, he repeatedly translates complex probabilistic reasoning into practical decision support for federal customers. He also brings proposal-writing and team leadership experience, coordinating multi-university and contractor efforts to deliver research-to-deployment outcomes. An unexpected thread through his career is a persistent interest in making arguable, teachable models—treating argument maps as tree-shaped Bayes nets—to improve how teams reason under uncertainty.
12 years of coding experience
15 years of employment as a software developer
Ph.D., Cognitive Science, History & Philosophy of Science, Ph.D., Cognitive Science, History & Philosophy of Science at Indiana University Bloomington
BA, Interdisciplinary Thesis: Astro/Phys, Philosophy, Anthropology, History of Science & Technology, BA, Interdisciplinary Thesis: Astro/Phys, Philosophy, Anthropology, History of Science & Technology at University of Virginia
Contributions:1 release, 9 commits, 3 PRs in 4 years 6 months
resource-allocationallocationoptimal
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.