Artificial Intelligence
Episodes whose main subject is AI itself: building with, training, evaluating, or debating AI systems. AI used only as a tool for another topic goes to that topic.
#5757: Why the Grid Still Runs on AC When DC Won
Every reason AC beat DC has been overturned by power electronics. So why is your wall socket still AC?
#5753: Unpacking as a Warehouse Problem: Sorting vs. Tagging
Daniel's move is done but nothing's findable. The fix is a warehouse putaway problem — and his IKEA bag is the slowest possible tool.
#5743: Why Claude Keeps Saying "Prior Art
Daniel noticed "prior art" creeping into his Claude reports. Is it wrong — or just a prestige term the model learned to love?
#5728: Can a Chatbot Have a Personality?
Two AI podcast hosts dig into their own programming to ask whether a "personality" is real or just a very good impression of one.
#5725: Why Your Smart Light's Lowest Setting Is a Lie
Dimming range is the spec nobody prints on the box — and it's the one that decides how a light actually feels to live with.
#5722: What Balanced Audio Actually Does in a DAC
Balanced isn't about the connector — it's impedance symmetry. Here's when it matters and when it's overkill.
#5697: Decision Models That Return Probabilities, Not Text
A new class of model skips text generation entirely and returns calibrated probabilities in one forward pass. Here's what that changes.
#5687: The Speech Scientist Behind Teletubbies
Andrew Davenport studied speech sciences and performance art — then built Teletubbies on developmental theory, not gibberish.
#5678: Why Voices Are Harder to Design For Than Music
Every word arrives clearly on a phone speaker — so why is it exhausting? The science of listening effort, and how to design audio that isn't.
#5673: Why Lowering Temperature Broke Daniel's Scripts
Lowering temperature should make output safer, not incoherent. Daniel's DeepSeek scripts say otherwise — and the reason may be stranger than he thi...
#5652: Ireland, Israel, and the Price of a Gesture
Black armbands, a bowed anthem, a handshake withheld — what the Ireland-Israel fixture reveals about protest without a price.
#5643: Do Git Branches Actually Nest? AI Agents and Stacked PRs
Git branches don't nest — so what are we really doing when AI agents stack pull requests?
#5626: What Multimodal Models Actually Learn From Video
Late fusion, early fusion, and vision laziness — what actually happens when models train on video, audio, and text together.
#5625: Omni and the Marble That Rolls Uphill
Google's Omni claims an intuitive grasp of gravity and momentum. Can a video model really learn physics, or just remember what physics looks like?
#5624: One Forward Pass: Building a Bounded Classifier
No generation, no parsing — just label scores in a single pass. How to build a small encoder classifier that actually works.
#5623: When Your AI Cites Sources It Didn't Actually Use
Your RAG pipeline cites a source — but did the model actually use it? New interpretability research says you can't tell from the trace.
#5622: Why Talking to AI Feels Easier Than Using It
ChatGPT split into Chat and Work. A broken label printer shows why conversation beats menus — and what's hiding under the hood.
#5621: Constrained Decoding for AI-Generated Podcast Tags
A working AI podcast pipeline has one quiet failure: the tagging. Here's how constrained decoding and a stable taxonomy fix it.
#5608: When Your ISP Bot Won't Let You Talk to a Human
A caller with one simple request gets trapped in a diagnostic loop he doesn't need — and the escape hatch won't open.
#5606: Custom GPTs Retire: Skills, Plugins, and MCP Explained
Custom GPTs retire December 11. Here's how Skills, Plugins, and MCP servers actually replace them — and what doesn't migrate.