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
The One-Charger Dream: Specs, Trade-offs, and Reality 16.04.2026 26:11
Can one desktop charger replace all your bricks? We break down the key specs—total wattage, PD 3.1, PPS, and GaN—and reveal the unavoidable trade-off with proprietary fast-charging phones. A practical guide to cutting cord clutter. Episode #112999 — open it directly at myweirdprompts.com/112999
Typst vs. LaTeX: The AI-Ready Document Engine 16.04.2026 43:21
The quest for beautiful, automated document generation is heating up. With Typst's stable release and the rise of AI agent protocols like MCP, we examine whether this modern contender can dethrone the venerable but complex LaTeX. We break down the core features—from declarative styling to human-readable errors—that make a typesetting system truly great for both humans and AI, and sketch the bluepr...
How to Test an AI Pipeline Change 16.04.2026 27:06
Iteratively testing AI agent pipelines is slow, expensive, and noisy. This episode explores a systematic engineering alternative: defining deterministic checkpoints within your pipeline. We break down how to instrument these checkpoints, use fixed seeds for reproducible testing, and apply evaluation platforms to get precise, actionable feedback on any change—turning pipeline tuning from alchemy in...
Why AI Agents Get Three Steps, Not Infinity 16.04.2026 37:45
Most AI agent demos promise endless autonomy, but the real engineering happens in the guardrails. This episode breaks down the "three-round rule": what a "round" of tool use actually is, why three is the magic number, and the two catastrophic failure modes—infinite loops and cost explosions—that this simple cap prevents. We ground it in a real stack using DeepSeek with native tool calls, explainin...
Why Lithium-Ion Won (And What's Next) 16.04.2026 27:05
Lithium-ion batteries power our world, but their dominance wasn't a marketing win—it was a physics inevitability. This episode explores why lithium's position on the periodic table made it unbeatable for portable power and how three decades of incremental engineering squeezed out massive gains. We then look at the real engineering challenges behind the next frontiers: silicon anodes that swell lik...
Agent-to-Agent Protocols: What Actually Needs Standardizing 16.04.2026 33:36
Agent-to-agent communication is moving from research into production, but the protocols powering it range from elegant to alarming. This episode digs into what a real A2A standard needs to specify—and what it can safely leave to implementers. We break down session handling and task lifecycles, the state management problem that everyone underestimates, security and authorization challenges unique t...
How Incentives Shape AI Safety Research 16.04.2026 34:57
The AI safety landscape is defined not just by where researchers work, but by the conflicting incentives of vendor labs, independent evaluators, and government institutes. This episode maps the ecosystem and asks what it actually means to do safety work under commercial pressure. Episode #292665 — open it directly at myweirdprompts.com/292665
Building Custom Benchmarks for Agentic Systems 16.04.2026 28:58
Standard benchmarks optimize for comparability across models, not for the specific failure modes and decision architectures that matter in production agentic systems. This episode walks through the full lifecycle of building custom evaluations: decomposing your workload, defining failure taxonomies with domain experts, constructing rigorous test sets, evaluating trajectories (not just outputs), an...
Why Israel Excels at Defense But Fails at Housing 16.04.2026 28:45
Israel presents a striking paradox: nearly eighty years of military excellence, a globally competitive tech sector, yet chronic failures in housing affordability, education quality, and poverty reduction. This episode explores what structural differences explain why some domains succeed brilliantly while others persistently underperform—and what the successes might teach us about fixing what's bro...
Building the All-Whiteboard Room: What It Actually Costs 16.04.2026 31:24
Daniel wants to transform his room into a collaborative whiteboard environment. Starting with one oversized board covered in agentic AI workflow diagrams, he's now imagining walls, ceiling, and even furniture all writable. But what does this actually cost? We break down the real products in the market—from budget whiteboard paint to custom porcelain steel panels to frameless glass—explore installa...
Constitutional AI: Anthropic's Theory of Safe Scaling 16.04.2026 31:02
What is Constitutional AI, really? Beyond the PR, Anthropic has a specific theory of how to make powerful language models safer: replace noisy human feedback with AI self-critique guided by a written constitution of principles. But this raises hard questions. Does replacing human judgment with AI judgment just move the problem? And what does Anthropic's safety mission actually assume about the rac...
Whiteboard Markers: The Tool Everyone Ignores 16.04.2026 29:49
Whiteboard markers are invisible until they fail. This episode digs into the massive gap between commodity markers and quality alternatives—the difference between wet erase and dry erase technologies, why the board surface matters as much as the marker itself, and what you actually buy if you're stocking a serious workspace. We talk Neuland, Edding, Staedtler, and the environmental math of disposa...
When "Global" Recession Means Rich Countries Sneeze 16.04.2026 25:41
The Gulf States warn that an Iran-Israel conflict could trigger global recession. But what does "global" actually mean? This episode unpacks the mechanics of economic shocks, why some economies decouple during downturns others can't escape, and the uncomfortable truth: "global recession" is really shorthand for "rich economies are contracting." We explore how oil shocks become contagion, why Austr...
What Enterprise AI Pricing Actually Negotiates 16.04.2026 30:20
When large organizations deploy internal tools on top of Claude, GPT-4o, or other frontier models, what's actually on the negotiating table? It's not the 50% discounts that enterprise software buyers are used to. Instead, enterprises negotiate service level agreements, data privacy terms, priority routing, and capacity planning. This episode unpacks why AI API pricing works differently from tradit...
AI as Your Ideation Blind Spot Spotter 16.04.2026 28:04
Expertise narrows imagination. Cognitive entrenchment, functional fixedness, and availability bias lock experts into narrow solution spaces—and they feel thorough the whole time. This episode explores how large language models can function as ideation partners that map the edges of possibility your brain has trained itself to ignore. We dig into concrete prompting strategies: constraint-breaking p...
When More Frameworks Make Worse Decisions 16.04.2026 28:52
How do you make a big decision well? We trace the surprising history of the pro/con list back to Benjamin Franklin's "Moral or Prudential Algebra" (1772), then explore why it fails—and what modern research-backed frameworks do better. From the WRAP method to regret minimization to second-order thinking, we map the landscape of structured decision-making. But here's the catch: more frameworks don't...
Who Does Every Country Owe Money To? 16.04.2026 25:49
When a government runs a deficit, it issues bonds to finance the gap. But here's the puzzle: most countries are in debt at the same time, and they often hold each other's debt. So who is the global creditor? This episode unpacks the actual mechanics of sovereign debt—why it's fundamentally different from personal borrowing, how currency denomination changes everything, and why the entire system hi...
How AI Benchmarks Became Broken (And What's Replacing Them) 16.04.2026 25:40
AI labs announce breakthrough scores on benchmarks like MMLU and HellaSwag constantly — but how much do these tests actually tell us about real AI capabilities? This episode digs into the messy reality of AI evaluation: how benchmarks get contaminated by training data, why they saturate within years, what models are really learning when they ace them, and what newer approaches like SWE-bench and L...
What Jerusalem Actually Needs to Survive 15.04.2026 33:37
What does preparedness actually look like when you live in a place where sirens are real, earthquakes are overdue, and the power might go out for days? Two residents of Jerusalem build a practical emergency course from scratch—covering the mamad, trauma first aid, food and water storage, and power management. Not prepper theater. Skills that save lives. Episode #089892 — open it directly at myweir...
The Hidden Career of Search and Rescue 15.04.2026 29:53
Search and rescue sounds like a single job—find the lost person, bring them home. But it's actually four completely different career paths with distinct training pipelines, operational tempos, and cumulative costs on the people doing the work. This episode explores military combat SAR (Pararescuemen, Unit 669), civilian urban USAR under FEMA, volunteer wilderness rescue, and Coast Guard maritime o...
Metal at Forty Thousand Feet 15.04.2026 27:44
What would happen if you dropped the Wright brothers' aerodynamic knowledge into 1903 with a mission to reach forty thousand feet? The answer isn't "it would be hard" — it's a categorical no. This episode traces the hidden metallurgical constraints that made high-altitude flight impossible until decades later: the fatigue science needed for pressurized cabins, the low-temperature ductility of allo...
What IP68 Actually Means (And Doesn't) 15.04.2026 31:43
When you see "IP68" on a product box, what are you really buying? In this episode, we unpack the gap between what ruggedness certifications claim and what they actually test. From the hidden details buried in IP rating definitions to how manufacturers exploit vague MIL-STD-810 claims, we explore how to read a spec sheet like an engineer—and why flashlight standards got it right when everything els...
Memory Isn't One Thing: What Science Actually Knows 15.04.2026 32:42
Most people dramatically underestimate what normal memory looks like, and overestimate how much of it is genetic destiny. This episode breaks down the five distinct memory systems, what twin studies actually tell us about nature versus nurture, and why chronic stress damages the hippocampus in ways that are reversible. Then: the surprising truth about photographic memory, eidetic imagery in childr...
Who Actually Wants AI to Slow Down? 15.04.2026 34:43
AI has grown faster than any technology in history, but should it? A listener asks whether the pace should actually slow—citing two reasons: technical (context windows remain the bottleneck despite hype) and human (expertise can't accumulate when the frontier resets every six weeks). The conversation explores who genuinely shares this worldview. Anthropic is the obvious anchor, but they're not arg...
One Remote, Three Streams: Building a Sane Media Setup 15.04.2026 26:12
Managing YouTube, Netflix, and local Plex content across multiple locked-down devices is a recipe for complexity. In this episode, we explore what a genuinely maintainable media setup looks like—from choosing between Raspberry Pi, Fire TV, and Chromecast, to why HDMI-CEC almost works, to the honest truth about why you can't build one app that handles everything. The answer isn't smarter devices. I...
Similar podcasts
Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.