Principal Applied Scientist - TS/SCI

Vantor•Herndon, VA

2d•$140,000 - $205,700•Onsite

About The Position

Vantor is seeking a Principal Applied Scientist to serve as the scientific authority for the development, validation, and application of advanced analytical and simulation models supporting characterization of the Ubiquitous Technical Surveillance (UTS) environment. This individual will combine expertise in computational modeling, applied mathematics, physics, and data science to ensure that machine learning models, agent-based simulations, and analytical frameworks accurately represent real-world phenomena and produce scientifically defensible, operationally relevant results. The role is instrumental in delivering a transition-ready capability that enables commanders and operators to evaluate operational risk and improve mission survivability. The team will design, develop, validate, and demonstrate an integrated analytical capability supporting pattern recognition, anomaly detection, predictive analytics, agent-based modeling (ABM) or equivalent, and decision support in a Ubiquitous Technical Surveillance (UTS) environment. The effort will enhance Joint Force survivability by enhancing the characterization of the UTS threat and providing a mission-agnostic risk-management tool for commanders and operators to utilize across the spectrum of conflict. The enterprise will leverage this analysis and these tools to increase isolated personnel (IP) survivability by enhancing evasion training, planning, and execution, while bolstering force protection for PR task forces.

Requirements

US Citizen and must have an active TS/SCI clearance, with ability to obtain CI Poly.
12+ years of relevant experience.
Master's or PhD in physics, mathematics, statistics, operations research, engineering, science, or related discipline.
Strong technical leadership experience.
Extensive demonstrated knowledge and experience utilizing the following modeling techniques: process, predictive, physics-based, agent-based.
Significant current experience (i.e., within last 2 years) in modeling, using statistics to conduct predictive modeling and make estimates on future occurrences based on prior events, and using physics-based modeling to review sensor capabilities and predict sensor ability to perform tasks in a synthetic scenario
Experience working with diverse data sets from UTS relevant sources
Proficiency in Python and scientific computing libraries.
Experience AI/ML programs.
Ability to clearly communicate technical concepts and analytical results in writing and verbally.
Must be able to work on-site in Herndon, Va.

Nice To Haves

Experience with agent-based modeling.
Experience supporting operational environments and/or Government R&D projects.
published modeling reports and/or authored new procedures/methods to create innovative modeling approaches to solve complex issues
Experience modeling human behavior, adversarial systems, or complex adaptive systems.
Experience with reinforcement learning, Bayesian inference, Monte Carlo methods, or stochastic simulation.
Familiarity with explainable AI (XAI) and integrating scientific principles into machine learning workflows.
Experience developing digital twins or high-fidelity operational simulations.
Experience supporting personnel recovery, force protection, intelligence analysis, or mission planning.

Responsibilities

Serve as the scientific lead for computational modeling and simulation activities across the program.
Establish scientifically rigorous methodologies for representing UTS phenomena within analytical and simulation environments.
Ensure analytical approaches accurately reflect the underlying physical, operational, and statistical characteristics of the data.
Design, develop, and validate complex agent-based models representing UTS interactions and operational environments.
Develop agent behaviors, environmental conditions, interaction rules, and emergent system dynamics
Evaluate simulation outputs for realism, repeatability, and operational relevance.
Evaluate government-provided datasets to identify meaningful physical, temporal, behavioral, and operational relationships.
Develop scientifically valid approaches for feature engineering and diagnostic vector identification.
Collaborate with data scientists to ensure machine learning models leverage meaningful scientific features rather than spurious correlations.
Assess the quality, completeness, and limitations of available data.
Apply statistical, probabilistic, and computational methods to characterize uncertainty and system behavior.