Gemini 3 is a few days old and the massive leap in performance and model reasoning has big implications for builders: as models begin to self-heal, builders are literally tearing out the functionality they built just months ago... ripping out the defensive coding and reshipping their agent harnesses entirely.<br/>Ravin Kumar (Google DeepMind) joins Hugo to breaks down exactly why the rapid evolution of models like Gemini 3 is changing how we build software. They detail the shift from simple tool calling to building reliable "Agent Harnesses", explore the architectural tradeoffs between deterministic workflows and high-agency systems, the nuance of preventing context rot in massive windows, and why proper evaluation infrastructure is the only way to manage the chaos of autonomous loops.<br/>They talk through:<br/>- The implications of models that can "self-heal" and fix their own code<br/>- The two cultures of agents: LLM workflows with a few tools versus when you should unleash high-agency, autonomous systems.<br/>- Inside NotebookLM: moving from prototypes to viral production features like Audio Overviews<br/>- Why Needle in a Haystack benchmarks often fail to predict real-world performance<br/>- How to build agent harnesses that turn model capabilities into product velocity<br/>- The shift from measuring latency to managing time-to-compute for reasoning tasks<br/>LINKS<br/>From Context Engineering to AI Agent Harnesses: The New Software Discipline, a podcast Hugo did with Lance Martin, LangChain (<a href="https://high-signal.delphina.ai/episode/context-engineering-to-ai-agent-harnesses-the-new-software-discipline" class="linkified" target="_blank">https://high-signal.delphina.ai/episode/context-engineering-to-ai-agent-harnesses-the-new-software-discipline</a>)<br/>Context Rot: How Increasing Input Tokens Impacts LLM Performance (<a href="https://research.trychroma.com/context-rot" class="linkified" target="_blank">https://research.trychroma.com/context-rot</a>)<br/>Effective context engineering for AI agents by Anthropic (<a href="https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents" class="linkified" target="_blank">https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents</a>)<br/>Upcoming Events on Luma (<a href="https://lu.ma/calendar/cal-8ImWFDQ3IEIxNWk" class="linkified" target="_blank">https://lu.ma/calendar/cal-8ImWFDQ3IEIxNWk</a>)<br/>Watch the podcast video on YouTube (<a href="https://youtu.be/CloimQsQuJM" class="linkified" target="_blank">https://youtu.be/CloimQsQuJM</a>)<br/>Join the final cohort of our Building AI Applications course starting Jan 12, 2026 (<a href="https://maven.com/hugo-stefan/building-ai-apps-ds-and-swe-from-first-principles?promoCode=vgrav" class="linkified" target="_blank">https://maven.com/hugo-stefan/building-ai-apps-ds-and-swe-from-first-principles?promoCode=vgrav</a>): <a href="https://maven.com/hugo-stefan/building-ai-apps-ds-and-swe-from-first-principles?promoCode=vgrav" class="linkified" target="_blank">https://maven.com/hugo-stefan/building-ai-apps-ds-and-swe-from-first-principles?promoCode=vgrav</a> <br/><br/>This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit <a href="https://hugobowne.substack.com?utm_medium=podcast&#38;utm_campaign=CTA_1">hugobowne.substack.com</a>