Context and Memory
299 episodes
#6169: What It's Like to Be a Token Inside an LLM
A token's-eye view of an LLM's forward pass — no cone, no squeezing, just a hallway where every layer scribbles on you.
#6164: We Gave Each Host Its Own AI Agent
We split our two hosts into separate AI agents with private context windows. The bill came to 78 cents an episode.
#6112: The Interpreter That Doesn't Know Who's Talking
Voice-to-voice translation is basically solved. The part where the machine knows who's in the room and who holds the power? Nobody's shipped it.
#6087: Building a Podcast Graph That Doesn't Suck
Daniel wants to turn a 6,000-episode archive into a walkable graph. Turns out the hard part isn't the database — it's the pruning.
#6050: Codex vs Claude Code: What the Numbers Actually Count
Codex has more users, Claude Code has more revenue — both can be true. The difference is what each number is counting.
#6040: Claude Code Subagents Can Now Spawn Subagents
Claude Code subagents can now spawn their own subagents. Here's where the depth limits come from, and why nobody actually knows where they stop.
#6037: Gemini 4 Argon: Google's Million-Token Flagship
A close read of Gemini 4 Argon's model card — a million-token window, thinking levels, implicit-only caching, and a lot of blanks.
#6036: GPT-6 Astra: A Million Tokens, No Fine-Tuning
A million-token window, ten built-in tools, a 272K pricing cliff — and no fine-tuning. What GPT-6 Astra actually offers builders.
#6033: Progressive Skills Discovery: When Bigger Context Windows Aren't Enough
Bigger context windows were supposed to make skill discovery moot. Here's why the listing budget still bites at scale.
#6029: One Model, Two Modes: Does AI Consolidation Hold Up?
OpenAI hides a router, DeepSeek merged two modes, Anthropic removed the knob. Is the model landscape really consolidating?
#6025: When Should Claude Code Start a New Session?
Long context windows don't fix stale context — they hide it. Why task boundaries, not token counts, are the real problem.
#6024: What's Actually Inside the KV Cache?
890 bytes per token: why the KV cache is both what makes LLMs practical and what makes long context expensive.
#6012: Inside the Decode Loop: How LLMs Generate One Token at a Time
The model doesn't choose tokens or decide when to stop — the engine does. A step-by-step tour of one decode iteration.
#6011: Inside LLM Inference: Prefill, Decode, and the KV Cache
One 8K-token sequence on a 13B model eats 6.7GB of KV cache—while the weights are only 26GB. Here's what actually happens per request.
#5972: AI Meeting Agents Are Becoming Stand-Ins
Bots now book follow-ups, update CRMs, and attend on your behalf. Here's what's real, what's speculative, and where the etiquette is heading.
#5948: LangGraph, LangChain, and Deep Agents in 2026
Runtime, framework, harness — how three LangChain layers actually fit together, and what changed in 2026.
#5915: How Multi-Agent AI Simulations Keep a Shared World State
Concordia keeps the world in a dictionary an LLM writes to. Generative Agents keeps it in a spreadsheet. CAMEL keeps it in the chat log.
#5908: One Database Won't Fix Your Fragmented Stack
Rita has two employees, four apps, and one fear: which database wins when they disagree? Here's the field-by-field answer.
#5903: Video Tokens, Omni, and the Local VRAM Math
Video costs ~100 tokens per second. That single number explains why ChatGPT hides video upload and why local inference hits a KV cache wall.
#5812: One Number, Every Hard Question: Agent Session State
A shopping bot that holds one cart total turns out to contain every hard problem in agent state management.
#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.
#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.
#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.