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.
#2063: That $500M Chatbot Is Just a Base Model
That polite chatbot? It started as a raw, chaotic autocomplete engine costing half a billion dollars to build.
#2062: How Transformers Learn Word Order: From Sine Waves to RoPE
Transformers can’t see word order by default. Here’s how positional encoding fixes that—from sine waves to RoPE and massive context windows.
#2061: The Memory Bottleneck That Drives Attention Design
Attention is the engine of modern AI, but it’s also a memory hog. Here’s how MQA, GQA, and MLA evolved to fix it.
#2060: The Tokenizer's Hidden Tax on Non-English Text
Why does a simple greeting in Mandarin cost more to process than in English? It's the tokenizer's hidden inefficiency.
#2059: When Your AI Agent Runs Stale Code
npx is silently running old versions of your AI tools. Here's why your updates vanish into a cache black hole.
#2057: How Agents Break Through the LLM Output Ceiling
The output window is the new bottleneck: why massive context doesn't solve long-form generation.
#2056: Music as Language: The Architecture Behind AI Song Generation
A look at how AI music models use audio tokens, transformers, and diffusion to turn text into songs.
#2049: Why Your Brain Prefers Listening Over Reading
Audio learning taps into ancient brain wiring, offering relaxed alertness and better big-picture retention than reading.
#2047: Why Video Calls Feel Like a Workout for Your Brain
Remote work is draining our "social radar," but new science shows how to rebuild it.
#2046: The Cinema of Constructed Reality
We asked an AI to curate films about AI and reality, exploring the psychedelic overlap between machine hallucinations and human perception.
#2043: From Wrappers to State Machines
Skip no-code traps. Learn the real stack for building agentic AI: Python, TypeScript, and Rust.
#2042: Gifted, Stigmatized, and Seeking Real Community
Why do online communities for the gifted become toxic, and how can you find real-world connections?
#2041: The "MPEG Moment" for AI: Llamafile & Native Models
Why are we squeezing massive cloud models onto desktops? Meet the "native" AI revolution.
#2040: The Rebellion Against Big Tech's AI Lock-In
Why run LLMs locally? We break down Ollama, llama.cpp, vLLM, and llamafile—and when to use each.
#2039: CLIs vs. MCPs: How AI Agents Actually Talk to Services
Why give an AI agent a terminal? We compare CLIs and MCPs for AI integration.
#2038: UI-First vs Architecture-First: Choosing Your AI Agent
LobeHub vs. Dify vs. n8n: We break down the chaotic landscape of local AI agents to find the right "brain" for your workflow.
#2037: The Hidden Hierarchy of Claude Code Extensions
Stop manually typing slash commands. Here’s the definitive hierarchy of Claude Code extensions—from legacy shortcuts to autonomous agents.
#2028: Agent Skills Are the New Apps
AI agents are getting an App Store for brains. Discover how modular skills are replacing massive prompts and what it means for the future of work.
#2027: The Missing Photoshop for Words
Why is editing text with AI so clunky? We explore the "TITO" paradigm—using small, local models for fast, private text transformation.
#2026: Prompt Layering: Beyond the Monolithic Prompt
Stop writing giant, monolithic prompts. Learn how to stack modular layers for cleaner, more powerful AI applications.