{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/69ab3b7c7036d739021982df/6a0c9d9fb23d275dceaa516d?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"I Asked Seven Questions About Our AI Agent. We Failed Five.","description":"<p>What's really happening inside the AI agent stack as agents move into production? The common story is that OpenAI and Anthropic decide whether your agent ships — but the reality is more complicated.</p><p><br></p><p>In this podcast, I share the inside scoop on the infrastructure companies quietly deciding whether AI agents reach production:</p><ul><li>Why runtime, identity, and data are the real control layers</li><li>How Cloudflare, Auth0, and Snowflake gate agent deployment</li><li>What separates a kill switch from telling the model to stop</li><li>Where Stripe and the card networks are racing on payments</li></ul><p><br></p><p>For builders and operators, the agentic shift is a real opportunity, but only if you map runtime, identity, data, payments, and observability for each workflow before it ships, not after.</p><p><br></p><p>Subscribe for daily AI strategy and news. For deeper playbooks and analysis: <a href=\"https://natesnewsletter.substack.com/\" rel=\"noopener noreferrer\" target=\"_blank\">https://natesnewsletter.substack.com/</a></p>","author_name":"Nate B. Jones"}