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.
#2684: When Agent Skills Collide: Context Windows & Plugin Design
How to handle overlapping agent skills and whether context windows will ever make the problem go away.
#2683: MCP vs Agent Skills: Context Wars
When 12M token windows arrive, do MCP servers or agent skills win? Plus: federated access for agent teams.
#2682: Live Retrieval vs. RAG: What an Agent Actually Does
Does every AI conversation create a tiny vector store? We unpack the real tradeoffs between live document fetching and pre-indexed RAG.
#2677: Memory Layers for AI Agents: SaaS vs Self-Hosted
Zep, mem0, Letta, Graphiti, Cognee — which memory layer should you commit to for your AI agent?
#2676: Vector Database Schema Design for AI Memory Layers
Stop dumping vectors blindly. Design metadata schemas and namespaces for retrieval that actually works at scale.
#2675: When AI Makes Documentation Effortless
The key documents every consultant needs—and how AI makes them effortless to create and maintain.
#2674: Why Your Agent's Context Window Is Getting Eaten Before You Start
Stop shipping the whole toolbox to every session. A bridge plugin pattern that fetches skills on demand instead.
#2673: The Embedding Coupling Problem: Editing Vector Stores
Can you edit or delete individual chunks in Pinecone? And can you actually back up a vector index? Yes—but with critical caveats.
#2672: When a Startup Claims to Break the Quadratic Wall
A startup claims linear attention scaling at 12M tokens, beating GPT-5.5 on retrieval benchmarks.
#2668: When a Flamethrower Is Overkill
Tesseract, EasyOCR, or a cloud vision model? How to build a fast, reliable label scanner for real-world conditions.
#2665: Partner Certs vs Personal Certs: What Actually Matters
Solo operators face structural barriers in vendor partner programs. Here's how personal and partner certifications actually differ.
#2664: Can You Trust an LLM's Raw Knowledge?
Why pre-trained knowledge isn't reliable for facts — and what actually makes models useful.
#2660: The Craft Cider Revival and the Art of Keeving
From Normandy's keeved ciders to Asturian sidra that argues with you — a global tour of craft cider's real hotspots.
#2657: When Puppeteers Stopped Hiding
Background removal isn't magic — it's multiple AI systems working in sequence. Here's what's actually happening under the hood.
#2653: The Hidden Infrastructure of American Puppetry
Tracing the surprising institutional depth of American puppetry, from UConn's puppet arts program to the Henson revolution.
#2651: AI Training Itself: Student, Teacher, and Grader
Can models generate their own training data and judge their own outputs? The promise and pitfalls of fully AI-led pipelines.
#2650: How to Catch an LLM's Bad Writing Habits
A practical guide to analyzing podcast transcripts for repetitive language and dialogue patterns — from Python word counts to embedding clustering.
#2648: The Art of the Brief: Writing What Busy People Actually Need
Why a crisp 600-word brief is harder than a 10-page report—and how AI changes the game.
#2647: Enforcing Async Boundaries When Clients Call Anytime
How to choose the right async tool, set boundaries with clients, and where AI fits in.
#2645: How to Document Failures for Your AI Second Brain
Stop writing diary entries. Start writing retrospectives your AI agent can actually use.