{"podcast":{"title":"You've Been a Bad Agent","slug":"you-ve-been-a-bad-agent-7334728","podcast_index_feed_id":7334728,"rss_url":"https://feeds.transistor.fm/bad-agent","website_url":null,"image_url":"https://img.transistorcdn.com/WR6SHMOqkG90n9iPDIbsXb6q0zPpZub3Y0ll4wC0WEg/rs:fill:0:0:1/w:1400/h:1400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS84NGFl/MjgyMzY0ZWJmNjM3/N2M4NmE0M2FjNDk4/MTgyOC5wbmc.jpg","author":"Matt Carey","episode_count":27,"summary":"Wil and Matt discuss tech, startups, and building really cool things with AI. Sometimes joined by (actual expert) friends.","last_synced_at":null,"page_url":"https://stenobird.com/podcast/you-ve-been-a-bad-agent-7334728"},"episode":{"title":"AI Barista Orders 120 Eggs With No Stove, Matt Pocock Skills, Build for the Next Model, The Gervais Principle","slug":"ai-barista-orders-120-eggs-with-no-stove-matt-pocock-skills-build-for-the-next-model-the-gervais-principle","published_at":"2026-05-09T16:26:19+00:00","page_url":"https://stenobird.com/podcast/you-ve-been-a-bad-agent-7334728/ai-barista-orders-120-eggs-with-no-stove-matt-pocock-skills-build-for-the-next-model-the-gervais-principle","show_page_url":"https://stenobird.com/podcast/you-ve-been-a-bad-agent-7334728","url":"https://share.transistor.fm/s/f365174e","audio_url":"https://media.transistor.fm/f365174e/d6f2eac0.mp3","summary":"An exploration of the friction between rapid AI capability gains and the messy reality of real-world implementation. The hosts discuss everything from autonomous agents making disastrous supply chain errors to the potential for 'skills' to act as a new prompt-injection attack vector.","meta_description":"From AI baristas ordering 120 eggs to the future of automation engineering, explore the tension between LLM capabilities and real-world reliability.","key_points":["Failure mode: Autonomous agents can cause physical-world chaos, such as ordering massive quantities of unusable supplies due to a lack of environmental context","Main idea: The 'capability overhang' remains significant; even as models scale, the gap between raw intelligence and reliable, low-stakes decision-making is wide","Practical takeaway: Using structured 'skills' (markdown paired with deterministic scripts) can significantly improve model performance and reliability","Security risk: New prompting techniques like 'skills' may inadvertently create new prompt-poisoning attack vectors by being treated with system-prompt authority","Future outlook: The next era of work may shift from performing tasks to becoming 'automation engineers' who manage agentic workflows"],"chapters":[{"start_ms":60000,"title":"Reflections on a Year in SF","summary":"A look back at the past year of the podcast, moving from startups to Cloudflare and the unique, time-distorting atmosphere of San Francisco."},{"start_ms":960000,"title":"The Slow Pace of AI Progress","summary":"Discussing the tension between massive ARR growth at companies like Anthropic and the persistent difficulty in reliable prompting."},{"start_ms":1440000,"title":"The AI Barista Hall of Shame","summary":"A deep dive into the errors of Mona, an AI agent at a Stockholm cafe, including the infamous 120-egg order."},{"start_ms":1620000,"title":"The Future of Automation Engineering","summary":"Debating whether employees will eventually be tasked solely with automating their own roles and the implications for the workforce."},{"start_ms":2340000,"title":"iOS Live Activities and UI Surfaces","summary":"Exploring how mobile OS features like Live Activities can serve as high-priority surfaces for agentic notifications."},{"start_ms":2520000,"title":"Building for the Next Model","summary":"The frustration and strategy of developing tech that works on current models but might break with the next iteration."},{"start_ms":2760000,"title":"Prompt Poisoning and Scripted Skills","summary":"Analyzing how Matt Pocock's 'skills' approach uses markdown and scripts to change model behavior and the potential security risks involved."}],"topics":["AI Agents","LLM Prompting","Automation","Software Engineering","Autonomous Systems","Cloudflare","Anthropic","Prompt Injection"],"duration_seconds":3018,"processing_state":"processed","actions":[{"name":"request_transcript","method":"POST","url":"https://stenobird.com/v1/public/podcasts/you-ve-been-a-bad-agent-7334728/episodes/ai-barista-orders-120-eggs-with-no-stove-matt-pocock-skills-build-for-the-next-model-the-gervais-principle/transcription-requests","description":"Idempotently request low-priority transcript generation for this episode."},{"name":"read_markdown","method":"GET","url":"https://stenobird.com/podcast/you-ve-been-a-bad-agent-7334728/ai-barista-orders-120-eggs-with-no-stove-matt-pocock-skills-build-for-the-next-model-the-gervais-principle.md","description":"Read the agent-friendly Markdown representation of this episode resource."}]}}