Job Description
CHAOS Industries is redefining modern defense with a multi-product portfolio that gives the ultimate advantage—domain dominance. The company's products are powered by Coherent Distributed Networks (CDN™), empowering warfighters, commercial air operators, and border protection teams to act faster, adapt rapidly, and stay ahead of evolving threats. CHAOS Industries was founded in 2022 and has raised a total of $1 billion in funding from leading investors, including 8VC, Accel, and Valor Equity Partners. The company is headquartered in Los Angeles, with offices in Washington, D.C., San Francisco, San Diego, Seattle, and London. For more information, please visit www.chaosinc.com . About the Team: Mission Engineering at CHAOS turns simulation output into decisions. We run large-scale modeling and simulation campaigns across all warfighting domains and the full kill chain, against named threats, in operationally relevant scenarios, at the speed engineering, operational, and customer teams actually need. Every CHAOS engineering trade, pursuit, and customer engagement is anchored in rigorous, physics-based, tactically relevant, and statistically valid analysis, and we're scaling the function to meet that bar across a growing product portfolio. About the Role: You will own statistical methodology for the Mission Engineering team at CHAOS. You'll design experimental constructs that extract meaningful signals from broad trade studies and computationally expensive simulation runs, build the analytical pipelines the team relies on, push the methodological state of the art on how we characterize uncertainty, build surrogate models, and communicate quantitative results to decision-makers. You will work shoulder-to-shoulder with engineers and experts in every domain to ensure that simulated, experimental, and tactical results presented by CHAOS are rigorous, reproducible, and actually deliver answers that our teams, customers, and partners need This is a foundational hire. You will have the freedom to move fast and set the standards for how CHAOS does quantitative analysis from day one. What You'll Do Design rigorous experimental constructs (DOE, space-filling designs, adaptive sampling, sequential experimentation) for large-scale simulation campaigns, getting maximum signal per simulation hour across operationally relevant trade spaces. Apply advanced statistical methods (such as regression modeling, Bayesian inference, surrogate/metamodeling, sensitivity analysis, uncertainty quantification, and beyond) to simulation output to produce decision-quality conclusions. Build and own scalable Python-based data pipelines for ingestion, processing, statistical analysis, and visualization of large simulation datasets. Develop ML and statistical surrogate models that accelerate analysis, enable real-time trade studies, and feed mission planning applications. Set team standards for data management, reproducibility, and statistical rigor (such as code review, methodology validation, and documentation practices). Translate operational and engineering questions into well-structured analytical approaches alongside M&S engineers, threat SMEs, and program staff. Push back when the framing is wrong. Author technical reports and briefing materials with clear, honest data visualizations; present quantitative results to senior technical and non-technical audiences in language they can act on. Mentor peers and cross-functional teams on experimental design, statistical methodology, and reproducible analysis. Support programs spanning DoD services, DARPA, intelligence community, and commercial customers. Required Qualifications Bachelor's degree or higher in Statistics, Data Science, Mathematics, Artificial Intelligence, a related quantitative field, or equivalent demonstrated expertise in modern statistical methodology. 7+ years applying advanced statistical and data science methods, ideally supporting defense, intelligence, or advanced technology programs. Deep working expertise in experimental design, regression and Bayesian methods, uncertainty quantification, and surrogate modeling, not just textbook familiarity. Strong proficiency working in Python, including scientific computing and ML libraries (especially Pandas, Polars, NumPy, SciPy, Scikit-Learn, Statsmodels, PyMC, Matplotlib, Seaborn, CuPy, PyTorch), and exposure to MATLAB or R. Demonstrated experience building scalable analytical pipelines for large datasets, including comfort with terabyte-scale data and modern dataframe tooling. Exceptional data visualization skills and the ability to develop briefing-quality technical products.