Develop and validate astrodynamics models across LEO, GEO, cislunar, and multi-body regimes. Generate synthetic observation datasets for machine-learning systems, implement orbit-determination and estimation chains, analyze covariance, observability, and estimator convergence, and define anomalous dynamical behavior. The engineer will collaborate with data science, software, systems engineering, and research-institution partners while contributing physics-based inputs to SDA programs and serving as the technical authority for orbital mechanics.

Develop machine-learning models and analytical pipelines for space domain awareness, including trajectory classification, anomaly detection, benchmarking, calibration, and evaluation under sparse and noisy observations. Integrate models with astrodynamics and embedded systems, optimize inference for constrained onboard hardware, and support model lifecycle, retraining, systems engineering, and operational documentation.

Profiles analytical workflows for deployment on spaceflight and resource-constrained platforms. Measures throughput, memory, latency, and power; identifies bottlenecks; evaluates CPU/FPGA partitioning and acceleration; and assesses algorithm simplification, radiation mitigation, and onboard architectures. Produces defensible trade studies, calibrated confidence metrics, and technology-maturation roadmaps. Collaborates with data science and astrodynamics teams on deployable models, hardware-in-the-loop validation, and spaceflight transition pathways.