RAG
99 episodes
#6146: Grounding Is Not Solved: DIY vs Gateway vs Native Search
Bolt-on search, gateway search, or native grounding? The plumbing didn't vanish — it just moved to someone else's bill.
#5950: Self-Hosting Your Search Stack: SearXNG, Crawl4AI, and the Real Bill
Can SearXNG, Crawl4AI, and an open-source aggregator really replace a commercial search API? Yes — with a very large asterisk.
#5949: Inside Deep Research Agents: Claims, Citations, and the 84.7% Orchestrator Problem
Why 84.7% of citation errors originate at the orchestrator — and how a claim ledger fixes it.
#5869: Scraping AliExpress Specs: Selectors, Agents, or Neither?
Daniel wants to snapshot AliExpress spec sheets into his inventory system. The selector approach is already dead — here's what actually works.
#5777: Lorebooks: The Hobbyist RAG That Beat the Labs
How a chat frontend fork built a full retrieval system — keywords, vectors, recursion — so fictional characters would stop forgetting who they are.
#5770: Talking to Your Data: What MCP Doesn't Solve
MCP standardized the pipe. It has no opinion about what flows through it — and that's where the hard part lives.
#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.
#5507: Inside ChatGPT's Hidden Pipeline: 8 Models Per Reply
One chat turn isn't one model call. It's moderation, PII filtering, a tone-based router, and activation classifiers — here's the documented graph.
#5230: Why Your Research Agent Won't Stop Searching
Most agent search runs add no new evidence — and 66-85% of the context window is just search snippets. Here's what actually stops the loop.
#5165: Weighting Memory in a RAG Pipeline
Daniel's memory leg won't fire. Turns out system prompts have no inherent vector-space privilege — but retrieval does have a real knob.
#4832: Your AI Memory Should Be Markdown Files
Why your AI project memory should live in Git-tracked Markdown files, not a proprietary platform.
#4829: Why Can't You Filter by Bluetooth on Amazon?
Technical buyers waste hours hunting specs. AI might finally fix e-commerce filtering.
#4809: Can a Podcast Train Itself?
Can a podcast become a self-developing project? We explore RAG vs. fine-tuning for lore memory.
#4194: Custom GPTs: When to Build, When to Skip
A decision framework for knowing when a custom GPT is worth the overhead — and when to just open a fresh chat.
#4106: Embedding Models vs LLMs: What Actually Connects?
Can you mix any embedding model with any LLM? And why are new embedding models still dropping if they're "solved"?
#4092: How AI Remembers What You Never Told It
How ChatGPT connected "wall anchors" to a power tool you bought days ago — without being asked.
#3751: Source-Restricted vs. Open Retrieval: How to Lock Down Your LLM
When should an LLM be locked to specific documents, and when should it search the web? A practical framework for grounding decisions.
#3120: What Makes Agentic Search Tools Like Exa Actually Work?
Why swapping Google for Exa transformed our show's accuracy — and what agentic search does differently.
#2705: Your Brain Isn't a Hard Drive — What Actually Fits
Long-term memory isn't storage — it's a generative model. Here's where the brain/computer analogy actually holds up.
#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.
#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.
#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.
#2664: Can You Trust an LLM's Raw Knowledge?
Why pre-trained knowledge isn't reliable for facts — and what actually makes models useful.
#2639: The Hidden Layer That Makes Search Work
Why your search results miss the mark — and how cross-encoders fix it.