Daniel Rosehill
My Weird Prompts
The human-AI collaboration podcast. A man, a sloth, and a donkey collaborate to create a podcast (with a little help from AI). No question is too obscure, no rabbit hole too deep. My Weird Prompts celebrates curiosity in all its forms. Daniel, the human, asks the questions that pop into his head at inconvenient moments. Corn the Sloth offers laid-back, thoughtful takes. Herman the Donkey brings boundless enthusiasm and energy. Together, they explore topics ranging from the mundane to the mind-bending. Each episode begins with a real voice memo from Daniel, processed through an AI pipeline that...
Be sure to visit the podcast's website and support the creator: myweirdprompts.com
Author
Daniel Rosehill
Category
Podcast website
Latest episode
Nov 21, 2026
Where to listen?
Podcasts in the app Replaio Radio Coming soonPodcasts are coming to the app soon. Install now and be the first to see a whole new take on podcasts
Episodes
Knowledge Without Tools: Why MCPs Aren't Just for Execution 13.04.2026 26:22
Most MCP coverage focuses on tools and execution, but the protocol's three primitives include Resources and Prompts—and a fully compliant MCP server can expose zero tools. This episode explores why you'd build a knowledge-only MCP instead of a REST API or RAG system, how to ground agents in authoritative sources like open government data, and what makes the MCP Resources primitive genuinely differ...
April Twenty-First: Israel's Ceasefire Collapse Moment 13.04.2026 24:58
With Iran's ceasefire expiring on April twenty-first—Israel's Memorial Day—military planners are signaling unprecedented readiness through strategic leaks to Hebrew media. The IDF has destroyed seventy percent of Iran's missile launcher arsenal, but the remaining thirty percent is the hardened, dispersed capability Iran protected most carefully. Meanwhile, Netanyahu's televised address about enric...
The UK's Impossible Choice in Trump's Iran War 13.04.2026 23:35
When the US escalated military action against Iran, the UK's response exposed a fundamental realignment in transatlantic relations. Under Starmer, Britain has refused base access, stayed out of blockades, and assembled a rival 40-country coalition—moves that invert the 2003 Iraq War dynamic when the UK sided with America. But the UK's tortured, incoherent position reveals something deeper: post-Br...
Reading the Geopolitical Forecast in Oil Prices 13.04.2026 24:42
The U.S. Navy blockade of Iranian ports just went live. Oil spiked. But here's the puzzle: WTI is still below April's highs, which means the market saw this coming. So how do you extract a real geopolitical forecast from commodity futures, options volatility, and prediction markets? This episode breaks down the three layers of market signals—the futures curve shape, options skew, and physical mark...
Mining the Strait: Why Clearing Iran's Weapons Takes Months 13.04.2026 26:38
When Iran mined the Strait of Hormuz, it did so so haphazardly that Iranian officials can't say exactly where the mines are. Now the US Navy faces an unprecedented challenge: clearing sophisticated acoustic and magnetic mines from a narrow, heavily defended shipping corridor without maps, without Iranian cooperation, and without enough minesweepers. This episode explores the technical complexity o...
The Strait Choke: How Naval Blockades Actually Work 13.04.2026 28:38
Trump announced a blockade of Iranian ports effective immediately, controlling a strait that handles 20% of global oil supply. But what does a blockade actually mean under international law? We dig into the history of naval blockades as a military tactic—from Dutch sieges in the 1600s to the Cuban Missile Crisis—and examine why some blockades (Japan in WWII) decisively ended conflicts while others...
Who Controls the Press Pool? 13.04.2026 25:10
The thirteen journalists who travel with the US president on Air Force One represent a century-old compromise between security and press freedom. But when the White House started controlling pool access in 2025, it exposed a fragile institutional arrangement. This episode traces the history of the traveling press pool in the US and Israel, the paradoxes of logistical dependence, and why the abilit...
The Invisible Workforce Behind AI 13.04.2026 27:38
AI models depend on human data labelers, yet this work remains invisible. This episode explores the annotation industry's scale, tools, and the paradox of AI automating the very labor that trains it. Episode #489269 — open it directly at myweirdprompts.com/489269
Nash's Real Genius (And Why the Movie Got It Wrong) 12.04.2026 29:26
Most people's understanding of game theory comes from a single scene in A Beautiful Mind—and it's wrong in a very specific way. In this episode, we unpack what Nash actually proved versus what the film dramatized, trace the difference between Nash equilibrium and Nash bargaining solution, and follow those ideas forward through a real game theorist's PhD work on network routing to an AI startup in...
Game Theory for Multi-Agent AI: Design Better, Fail Less 12.04.2026 28:23
When you build multi-agent AI systems, you're designing a game—and if you don't understand game theory, you're designing it badly. This episode covers the foundational concepts that shape how AI agents interact: Nash equilibrium, dominant strategies, zero-sum versus positive-sum games, and the prisoner's dilemma. Then it pivots to the practical toolkit: mechanism design, incentive compatibility, a...
Running Claude in Your Apartment (The Physics Says No) 12.04.2026 27:01
What does it actually take to run a state-of-the-art coding AI locally? Corn and Herman spec out three tiers of hardware—from the "Reasonable Madman" build at $11K to the "Nuclear Option" at half a million dollars—and then confront the physics: 18,766 BTUs of heat per hour, 90 decibels of continuous noise, and the thermodynamic certainty that your apartment will become uninhabitable without interv...
How We Built a Podcast Pipeline 12.04.2026 29:00
For over two thousand episodes, the production pipeline has run invisibly—until now. In this rare technical deep dive, Hilbert walks through the entire system: how Daniel's late-night voice memos become polished scripts, why the pipeline switched from Gemini to Claude Sonnet 4.6, how prompt caching cut costs by ninety percent, and what three A10G GPUs do during voice generation. Learn about LangGr...
Making Multi-Agent AI Actually Work 12.04.2026 24:29
The AI industry is building complex multi-agent systems at scale, but the people actually shipping them are quietly saying you probably don't need them. We dig into the empirical case against multi-agent architectures—including a Google DeepMind study of 180 agent configurations, Stanford's mathematical proof that single agents outperform on reasoning tasks, and direct admissions from Anthropic an...
Simulating Extreme Decisions With LLMs 12.04.2026 23:30
The CIA's operational assessment of Snow Globe—IQT Labs' AI wargaming platform—alongside a Stanford and Hoover Institution study of 214 national security experts reveals a structural problem: large language models cannot faithfully simulate extreme human decision-making. When assigned personas as pacifists or sociopaths, GPT-3.5, GPT-4, and GPT-4o produce statistically indistinguishable outputs. T...
Scaling Multi-Agent Systems: The 45% Threshold 12.04.2026 25:15
Everyone's building multi-agent systems. But a new Google DeepMind and MIT paper tested 260 configurations across six benchmarks and found something counterintuitive: independent agents amplify errors 17x compared to single agents, every multi-agent variant degraded sequential reasoning by 39-70%, and coordination overhead costs 1.6-6x more tokens for matched performance. The research reveals a cl...
Is Emergence Real or Just Bad Metrics? 12.04.2026 24:22
When models scale up, do genuinely new capabilities suddenly appear—or are we just measuring improvement badly? This episode digs into the Wei et al. emergence paper, the Schaeffer et al. rebuttal that called it a "measurement mirage," and where the science actually stands. We cover the mathematical argument behind metric artifacts, the cases emergence skeptics can't explain away (like chain-of-th...
Why Claude Writes Like a Person (and Gemini Doesn't) 12.04.2026 26:42
Why does Claude produce writing that sounds like an actual person, while Gemini—despite being genuinely impressive at code, reasoning, and retrieval—generates text that reads like a very good search result? This episode works backwards from that observed quality gap to explore the mechanistic explanation: Constitutional AI versus standard RLHF, the "assistant-brained" problem, and why reasoning mo...
The AI Persona Fidelity Challenge 12.04.2026 27:33
The world's most capable language models can ace any standardized test, yet they routinely fail at one of the most humanly intuitive tasks: maintaining a consistent persona across a conversation. New dialogue-specific benchmarks and wargaming research reveal a striking gap: models playing strict pacifists and aggressive sociopaths show no statistically significant behavioral difference. We explore...
Taking AI Agents From Demo to Production 12.04.2026 30:22
Building an LLM agent that works in a notebook takes a day. Getting it reliable in production takes weeks. This episode unpacks the invisible infrastructure gap that tutorials skip: full-stack observability, prompt versioning as a safety problem, A/B testing with non-deterministic models, canary deployments, rollback strategies, and the human oversight question nobody wants to answer. We walk thro...
The Economics of Running AI Agents 12.04.2026 27:04
AI agents are bankrupting projects at scale. A single misconfigured agent loop can cost $47,000 in 48 hours, and 40% of agentic AI projects fail due to hidden costs. This episode breaks down the engineering playbook for production cost control: dynamic model routing across capability tiers, prompt caching strategies that differ by provider, token budget allocation by priority instead of chronology...
Making Voice Agents Feel Natural 12.04.2026 28:37
Voice transcription and synthesis sound great, but talking to a voice agent still feels slightly off. Why? Because the hard problems are invisible: how agents detect when you've actually finished speaking versus just pausing to think, how they handle interruptions without cutting you off mid-sentence, what happens when latency budgets blow, and whether they can read emotional tone. This episode di...
Can You Actually Review an AI Agent's Plan? 12.04.2026 25:39
AI agents are getting smarter at planning, but there's a critical gap between having a plan and letting humans see and approve it before anything breaks. This episode digs into ReAct, plan-and-execute, ReWOO, tree-of-thought, and Reflexion—the five major planning patterns reshaping how agents reason. We explore why most agents today hide their plans in context windows or internal reflections, how...
When RAG Becomes an Agent 12.04.2026 29:04
Retrieval-Augmented Generation looks straightforward in a chatbot: query, retrieve, answer. But inside an AI agent, it becomes something fundamentally different — a loop with decision points, multiple knowledge sources, and the ability to refine, evaluate, and even write back to its own knowledge base. This episode breaks down five core architectural differences that separate agentic RAG from the...
The Sandboxing Tradeoff in Agent Design 12.04.2026 31:47
Giving AI agents tools to execute code, write files, and make API calls creates a fundamental tension: sandboxing them makes them useless, but leaving them unrestricted invites catastrophe. This episode breaks down the containment paradox that researchers have identified as unsolvable—you can only manage it. We cover the major isolation approaches (E2B, Daytona, Modal, Firecracker microVMs, Docker...
Building Cost-Resilient AI Agents 12.04.2026 35:24
AI agents sound cheap until they fail. A single fifty-turn session costs ninety cents—but when agents loop or restart from scratch after a mid-workflow failure, that cost multiplies fast. An eighty-five percent reliable step sounds solid until you compound it across ten steps: you're down to twenty percent success. This episode digs into the engineering that prevents wasted money when agents break...
Similar podcasts
Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.