Ruze Richards

About

I’m an AI and software architect with 20+ years of experience designing, building, and leading complex, mission-critical systems. My current focus is the architecture of production-grade AI systems: LLM applications, agentic workflows, memory and knowledge infrastructure, retrieval and reasoning systems, and the engineering required to make generative models reliable in real-world environments. My work increasingly focuses on the security and reliability of AI systems, including adversarial prompting, agent safety, tool-use security, code generation risks, evaluation frameworks, model behavior analysis, and the secure deployment of production AI applications. I regularly evaluate frontier models for software engineering, security analysis, incident investigation, and autonomous agent workflows. I work at the intersection of AI research and production engineering turning current AI capabilities into systems that are observable, scalable, maintainable, and commercially useful. This includes designing multi-agent systems, long-term memory architectures, knowledge graphs, evaluation frameworks, and hybrid AI systems that combine models with deterministic software, tools, data, and busi

Experience

Education