Joshua Bone

Background

About Joshua Bone

As a nuclear-trained submarine officer turned software engineer, I bring a rare blend of mission-critical operational leadership and large-scale software engineering.I spent 10 years as a U.S. Navy submarine officer on fast attack submarines, serving as a watch team leader and as a Division Officer (Machinery Division, Nuclear). In those roles, I operated in high-stakes, real-time environments where the margin for error is effectively zero. As the Commanding Officer’s direct representative on watch, I led a 25-person watch team responsible for safe and effective ship operations across sonar, weapons, navigation, damage control, atmosphere control, hydraulics/electrical, depth-keeping, and nuclear propulsion. I also led a 12-person propulsion plant watch team and owned the long-term health of a division of nuclear-trained personnel through maintenance execution, work controls, training, supply/inventory, and readiness. I held TS/SCI and SECRET clearances and learned firsthand what disciplined execution looks like when the system is a $3B asset and lives depend on it.Over the past decade in Big Tech (Google + Meta), I’ve applied that same rigor to building durable, auditable, high-throughput software systems—especially where correctness, traceability, and reliability matter as much as performance.At Google (GCP), I designed and maintained large-scale batch pipelines in Java/MapReduce to process and export transactions representing billions of dollars across Google Cloud customers. I built production features like append-only undo, stateful metadata caching, manual backfill, and release automation—while supporting oncall and contributing as a Java Readability reviewer and SWE interviewer (100+ candidates).At Meta, I led end-to-end development of a Python-based attribution rules framework covering four of the largest Monetization platforms (~95% coverage), consolidating ad-hoc SQL and spreadsheets into a consistent, auditable system. I architected a modular, component-based pipeline enabling parallelization, caching, and auto-generated documentation, cutting runtime from ~2 hours to ~5 minutes. I also built debugging tools that uncovered inconsistencies and saved hundreds of engineering hours, and I mentored and onboarded new engineers to improve team velocity and quality.