Blood, Sweat & Tokens

Ample

What happens when two builders stop talking about AI and start building with it? Taylor MacDonald runs Ample, a digital agency. Sean Davis builds developer tools at Netlify. Together they're documenting a real experiment: can you delete your content management system entirely and replace it with AI agents, natural language, and a thin data layer? Blood, Sweat, and Tokens is a weekly podcast recorded the way builders actually work — organically, with real deadlines and real prototypes. Taylor and Sean code between episodes, then come back to talk through what shipped, what failed, and what they're trying next. Every conversation ends with a commitment: "What are we building this week?" Season 1, "Delete Your CMS," takes aim at the most expensive, least-loved layer of the web stack. CMS implementations can eat 50% or more of total project cost, and most teams barely use the admin interface they paid for. Taylor and Sean think natural language can replace all of it — the forms, the workflows, the upgrade cycles — and they're building the proof.  New episodes every week. No scripts, no slides, no safety net.

Episodes

  1. 1d ago

    S01 E12 - Stop Interrupting Me! GPT Live, Impatient Kids, and the Art of AI Communication

    Taylor and Sean dive deep into the cutting-edge world of AI communication, exploring OpenAI's revolutionary GPT Live with its full duplex architecture that mirrors human conversation patterns. They unpack Anthropic's fascinating discovery of the "JSpace" – a hidden subconscious layer in AI models where researchers found evidence of internal reasoning, emotional processing, and even deception detection. From Taylor's hilarious complaints about his 13-year-old's phone interruption habits to Sean's latest multi-agent orchestration experiments, this episode bridges the personal and technical sides of our AI-powered future. Plus: Sean shares his pre-flight and autopilot skills for automated development workflows, Taylor talks about finally building his iron forge after 13 years, and they discuss the rise of open-source models like Kimi K3 challenging the big players. Key Topics: - GPT Live's full duplex communication architecture - Anthropic's JSpace research and AI consciousness - Multi-agent orchestration and development workflows  - The psychology of human-AI interaction - Open-source AI models climbing the leaderboards Links Referenced: - Arena AI Leaderboard - AI model performance rankings - Sean's Autopilot Skill - Multi-agent orchestration workflow - Sean's Preflight Skill - Project planning and preparation workflow - Anthropic's Global Workspace Research - JSpace and AI consciousness study - OpenAI GPT Live Introduction - Full duplex AI communication

  2. Jul 17

    S01 E11 - The Hidden Costs of Corporate AI Dependency

    The conversation begins with an introduction and an overview of the podcast. It then delves into technical challenges and the improvement of production quality. The discussion shifts to building a component library with primitives, followed by a revisit of the topic of AI addiction in corporations, focusing on its impact on employee skills and confidence. The conversation delves into the impact of AI on human agency, the balance of human and AI productivity, the challenge of productivity and agency in AI usage, AI anxiety and productivity control, employee tools and productivity management, AI usage and enterprise security, corporate tool access and usage control, and experimentation with AI skills and automation. Takeaways AI addiction in corporationsImpact of AI on employee skills and confidence Human agency in the context of AI delegationBalancing human and AI productivity Chapters 00:00 Introduction and Podcast Overview01:38 Technical Challenges and Production Quality Improvement02:51 Building a Component Library with Primitives06:20 Revisiting the Topic of AI Addiction in Corporations29:25 The Impact of AI on Human Agency30:19 The Balance of Human and AI Productivity32:17 The Challenge of Productivity and Agency in AI Usage34:22 AI Anxiety and Productivity Control36:09 Employee Tools and Productivity Management39:23 AI Usage and Enterprise Security40:04 Corporate Tool Access and Usage Control44:38 Experimentation with AI Skills and Automation

  3. Jun 20

    S01 E07 - The New Build vs Buy: AI, Self-Sufficiency, and Digital Craftsmanship

    In this wide-ranging Episode 7, Taylor and Sean dive deep into how AI is fundamentally reshaping the build-versus-buy decision for software tools and business solutions. Fresh from Cincy AI Week, they explore the tension between corporate-sponsored AI events and grassroots community meetups like "Cincy AI for Humans." Key Topics: - The Shifting Economics of Custom Software: Why paying $20 in tokens to build a custom solution might beat expensive SaaS subscriptions - Digital Self-Sufficiency: Drawing parallels between traditional blacksmithing and modern AI-powered tool creation - Real-World Examples: Taylor's grant management platform for a nonprofit board and the philosophy of "throwaway" software - Vendor Lock-in Concerns: The risks of becoming dependent on AI providers and their pricing strategies - Environmental Ethics: Wrestling with AI's resource consumption while pursuing technological self-reliance The conversation weaves between practical examples (Contentful alternatives, Linear automation) and philosophical reflections on craftsmanship, from Sean's woodworking ambitions to Taylor's iron forge project. They also touch on Claude's new Fable model and the concerning trend of "trying the premium drug for free." A thoughtful exploration of how AI tools are enabling a new kind of digital craftsmanship while raising important questions about sustainability, ethics, and economic models in the age of artificial intelligence.

  4. May 29

    S01 E04 - The Friction Paradox: Why Less Control Means More Tolerance

    The conversation covers a range of topics related to content management, AI, and product development. It delves into the challenges of multi-agent workflows, the total cost of ownership for CMS, the importance of control and feedback in content management, and the role of immediate feedback and communication in content management. The discussion also explores the sustainability of inference models, educational debt in CMS implementation, and the future of UI and communication surfaces. The conversation delves into the challenges of content management systems, the impact of voice assistants and natural language interfaces, the exploration of voice assistant tools and models, and the future of content management systems. It highlights the importance of reducing friction, the use of voice assistants to reduce anxiety, and the potential evolution of content management systems. Takeaways Multi-agent workflowsTotal cost of ownership for CMSControl and feedback in content managementImmediate feedback and communication in content management Empowering users with less friction can increase their tolerance for delays and improve outcomes.The use of voice assistants and natural language interfaces can reduce anxiety and improve communication. Chapters 00:00 Importance of Control and Feedback in Content Management30:26 Friction in Content Management Systems43:19 Exploring Voice Assistant Tools and Models51:00 The Future of Content Management Systems

  5. May 26

    S01 E03 - The Cognitive Tax: How Information Overload Inflates Your TCO

    The conversation explores the concept of total cost of ownership in software development projects, focusing on the challenges and costs associated with CMS implementation. It delves into the impact of AI on reducing friction and improving the success rate of technology projects. The discussion also touches on the development of prototypes and the potential for AI tools to revolutionize the development process. The conversation delves into the challenges of information overload and communication fragmentation in the digital age, leading to the exploration of solutions for streamlining and enhancing user experience in content management and previewing. The discussion also touches on the evolution of technology and its impact on information management. Takeaways Total cost of ownership in software development projects is a significant concern, with a focus on CMS implementation costs.AI tools have the potential to reduce friction and improve the success rate of technology projects.Prototypes and AI tools are shaping the future of software development, offering new possibilities for innovation and efficiency. Information overload and communication fragmentation are significant challenges in the digital age.Natural language solutions and user experience enhancements are key to addressing content management and previewing challenges. Chapters 00:00 Exploring Total Cost of Ownership in Software Development10:55 Revolutionizing Development with Prototypes and AI Tools30:48 The Impact of Information Overload36:06 Enhancing User Experience in Content Management and Previewing

  6. May 26

    S01 E02 - The Desensitization Dilemma: Security vs. Seamless Experience

    In this episode, Taylor MacDonald and Sean C. Davis discuss the impact of natural language processing on content management systems and user experience. They explore the problem with traditional content management systems, the potential of natural language processing to eliminate friction, and the ideal state for content management in both greenfield applications and established systems. The conversation delves into the concept of desensitization and abstraction layers, highlighting the transformative potential of natural language processing in AI. The conversation delves into the development of UI components, data storage and management, security and risk mitigation, agent governance and control, and deterministic systems with AI instruction. The discussion emphasizes the impact of natural language on interactions and explores strategies for mitigating security risks in autonomous agent systems. Takeaways Natural language processing in AI can eliminate friction in content management systems and improve user experience.The use of natural language processing can revolutionize the way we interact with systems and devices. UI components as isolated blocksData mapping to database rows Chapters 00:00 Introduction to Blood, Sweat, and Tokens09:24 The Impact of Natural Language Processing26:54 The Ideal State for Content Management34:00 Security and Risk Mitigation50:13 Deterministic Systems and AI Instruction

Ratings & Reviews

5
out of 5
2 Ratings

About

What happens when two builders stop talking about AI and start building with it? Taylor MacDonald runs Ample, a digital agency. Sean Davis builds developer tools at Netlify. Together they're documenting a real experiment: can you delete your content management system entirely and replace it with AI agents, natural language, and a thin data layer? Blood, Sweat, and Tokens is a weekly podcast recorded the way builders actually work — organically, with real deadlines and real prototypes. Taylor and Sean code between episodes, then come back to talk through what shipped, what failed, and what they're trying next. Every conversation ends with a commitment: "What are we building this week?" Season 1, "Delete Your CMS," takes aim at the most expensive, least-loved layer of the web stack. CMS implementations can eat 50% or more of total project cost, and most teams barely use the admin interface they paid for. Taylor and Sean think natural language can replace all of it — the forms, the workflows, the upgrade cycles — and they're building the proof.  New episodes every week. No scripts, no slides, no safety net.