Can Agentic Engineering Really Deliver Enterprise-Grade Code? | Reuven Cohen

In this episode of The AI Native Dev podcast, host Simon Maple interviews Ruben Cohen (Ruv), an agentic engineer and founder of the Agentix Foundation and Claude Flow. They discuss the current state of agentic coding, its evolution, and the challenges developers face, including the high failure rate of agentic projects due to lack of expertise. Ruv explains the Agentix Foundation's mission to standardize agentic engineering and introduces Spark, a specification-based approach for AI implementation, testing, and verification. He also details Claude Flow, a tool he created to enable parallel agentic development using Claude Code, emphasizing the importance of specifications, pseudocode, and test-driven development. The conversation touches on the balance between human oversight and AI autonomy, the significance of validation in agentic workflows, and the future of agentic software development.