Nvidia CEO Jensen Huang on the future of coding: “ Every engineer is going to have 100 agents.”

In this episode of All-In Podcast, the four besties dive deep into a conversation about the future of software engineering and artificial intelligence. The episode centers on a provocative thesis: every engineer will soon have access to 100 AI agents that will fundamentally reshape how code gets written and software gets built. This represents a seismic shift in the profession, moving away from engineers manually writing code line by line toward a model where human developers act as architects and directors of autonomous AI agents. The hosts explore what this means for the tech industry, the job market, and the next generation of engineers entering the field. They discuss how Nvidia's infrastructure and GPU technology are positioned to power this agent revolution, enabling the massive computational resources required to run hundreds of AI agents simultaneously across enterprises. The conversation touches on the democratization of software development that could result from this shift. If every engineer, regardless of skill level, can leverage 100 capable AI agents, the barrier to entry for building sophisticated applications drops dramatically. This could unleash a wave of innovation from engineers and builders who previously lacked the time or expertise to implement complex features. The besties examine the transition period and potential disruption this will cause. Current senior engineers who've built expertise over decades may see that expertise become less relevant if routine tasks are automated. Conversely, younger engineers entering the field will never know a world without AI agents and may develop fundamentally different problem solving approaches. They debate whether this is a net positive for the industry and society. The hosts also consider the implications for startup formation and entrepreneurship. If AI agents can handle much of the heavy lifting on software engineering, then teams can ship products faster with fewer people. This could accelerate the pace of startup formation but also create new competitive dynamics. The conversation explores risks including potential security vulnerabilities in agent-driven development, the challenge of managing and training these AI systems, and questions about code quality and reliability. Throughout the episode, the hosts maintain their signature blend of optimism about technological progress and healthy skepticism about unforeseen consequences and transition challenges.