That covers motor architecture, grain design, internal ballistics, performance prediction, igniter design, case and nozzle integration, and the work of tying predicted, ground-tested, and flight-test motor behavior together so the model keeps getting better after every fire. AI-native working style: daily use of agentic coding tools (Claude Code, Codex, OpenCode, or similar) for ballistic-code scripting, performance prediction, and post-test data reduction.