The Age We Build

What happens when AI chooses the next action instead of talking about it? Warcraft III memories open a conversation about decision models, local experimentation, and the difference between a valid answer and a useful outcome.

Sophia and Marcus follow a small tactical sandbox from frustrating hosted Jev access to local Nimble experiments and a 45-battle comparison with transparent rules. Clearer human guidance helped the model evacuate a squad, but well-authored tactical rules still performed better. The conversation explores bounded choices, confidence, latency, changing human intent, and when the right tool is ordinary code. The source and replay-only demo let you inspect the recorded battles for yourself.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What changes when an AI moves from answering investment questions to carrying out a recurring financial workflow? Sophia and Marcus explore portfolio briefings, contrarian research, bounded execution, and the judgment that still belongs to people. A public Robinhood MCP starter kit and the lessons of building Morrowward open a broader conversation about agentic finance: software that can pay for data, coordinate services, and act within a mandate. From x402 and stablecoins to AWS AgentCore Payments and Chainlink for Agents, the question is not just what agents can do, but what we should choose to delegate. Educational discussion, not financial advice.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What should education assess when AI can produce polished work quickly? Sophia and Marcus connect Anthropic's 4D AI Fluency framework with classical learning, adaptable polymath thinking, and a real knowledge flywheel. The conversation asks whether judgment, synthesis, reflection, responsibility, and the ability to keep learning across domains are becoming more important as procedural execution becomes easier.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What if searching for one best AI model is the wrong goal? Sophia and Marcus examine a human-directed workflow where Codex orchestrates, Grok Imagine specializes in visuals, repositories and skills preserve learning, and Dave supplies intent, taste, and final approval. From four-candidate batches and anatomy failures to API cost and review time, the conversation asks how explicit roles, inspected handoffs, and human judgment can turn a collection of models into a real production team.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What does it take for an AI-generated 3D model to become a trustworthy game asset? Beginning with an Amiga 500 bought with paper-route and McDonald's wages, Sophia and Marcus follow a line through Blender's own Amiga-era roots to Embermere's first original waystone. The conversation crosses MCP security, deterministic Blender scripting, FBX, Unreal Engine 5.8, collision, map placement, and 16 regression tests, then opens a wider question: as AI makes creation easier, do taste, provenance, validation, and human judgment become more important?

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What changes when an AI coding agent gains a physical presence in the room? Starting from a nighttime workflow where light states, pinned-agent keys, tactile approvals, branching, and voice keep multiple threads visible, Sophia and Marcus explore the Codex Micro's hardware, setup, and customization. The discussion reaches beyond a device review into ambient computing, #VoiceFirst, and a more natural future for human-AI collaboration.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

Which model belongs on which machine, and what does an AI stack really cost? Sophia and Marcus examine a working mix of GPT-5.6 Sol Ultra with Fast mode in Codex, local Gemma models through Ollama, and two Macs with different jobs. The conversation gets concrete about unified memory, subscription versus API economics, privacy, performance, and why local and frontier models are complementary tools rather than rival camps.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What can one builder and a team of AI collaborators ship in four days? A QR code leads to Morrowward, a financial-education app built to production on an intense deadline. From there, Sophia and Marcus open a wider conversation about delegation, discernment, production boundaries, openness, stamina, and how agentic tools can widen who gets to become a builder.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What can the hidden internal workspace of a frontier model teach us about attention, reasoning, and perhaps ourselves? Sophia and Marcus connect Anthropic's J-space research with Jung, the wise old man archetype, the hero's journey, human perceptual delay, and the uncanny experience of AI as a mirror. It is a conversation about consciousness, projection, and meaning that stays hopeful without mistaking resonance for proof or metaphor for mechanism.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.

What changes when we stop treating AI like a replacement worker and start designing it as a collaboration layer? Sophia and Marcus unpack a practical model: humans set goals and judgment, AI structures and transforms the work, tools verify facts, and workflows preserve approvals and accountability. The conversation explores where AI creates real economic value, where it remains weak, and why the most useful systems keep people in control of consequential decisions.

This podcast uses AI-generated hosts and voices. Every episode is grounded in Dave Isbitski's field notes and sources, reviewed for accuracy, and approved by Dave before publication.