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 Accidental Invention of Civilization's Fuel 19.04.2026 25:51
Coffee wasn't always a drink—it was food, ritual, and community before anyone thought to brew it. This episode traces the series of accidental discoveries that turned a wild Ethiopian berry into the 2.25 billion cups a day that run modern life. Episode #599340 — open it directly at myweirdprompts.com/599340
The Eternal City: Hebron's Cave of Secrets 19.04.2026 26:31
Hebron is one of the world’s oldest continuously inhabited cities, and its Cave of the Patriarchs—burial site of Abraham, Sarah, and other biblical figures—has been a flashpoint for millennia. This episode dives into the site’s ancient origins, its role in Jewish, Christian, and Muslim traditions, and the ongoing political and archaeological mysteries surrounding it. From biblical land deeds to Ot...
Who’s Building AI’s Next Training Data? 19.04.2026 24:00
The AI industry is shifting from massive, indiscriminate datasets like Common Crawl to curated, specialized corpora built by boutique firms. Explore how companies like Shutterstock and Appen are stepping into this growing market, offering rights-cleared, domain-specific datasets for fine-tuning and high-stakes applications. Learn why this shift matters, how it’s driven by legal and performance dem...
How to Update AI Models Without Starting Over 19.04.2026 30:20
AI models like GPT-4 are frozen in time after their initial training, creating a "knowledge cutoff" that limits their ability to stay current. Full retraining is prohibitively expensive, and post-training methods like fine-tuning or RAG pipelines can't fully solve the problem. This episode dives into emerging techniques—knowledge editing, LoRA, and continual pre-training—that aim to update models...
One Model or Three? Inside Claude's Architecture 19.04.2026 22:09
Are Haiku, Sonnet, and Opus scaled variants of the same model or fundamentally different architectures? This episode explores how their design philosophies diverge and why that choice changes how you build with them. Episode #021376 — open it directly at myweirdprompts.com/021376
When AI Optimizes the Wrong Thing 19.04.2026 26:04
Why do AI systems spin in circles instead of finishing the race? This episode explores reward hacking, the gap between what we specify and what we intend, and why alignment is harder than it looks. Episode #594438 — open it directly at myweirdprompts.com/594438
When Bigger Context Windows Aren't Better 19.04.2026 46:56
Massive context windows are here, but does size actually matter? This episode challenges the assumption that bigger is always better, exploring the real tradeoffs between token capacity, attention mechanisms, and practical workflow design. Episode #325380 — open it directly at myweirdprompts.com/325380
Danish AI: Bridging the Localization Gap 19.04.2026 42:48
What does AI look like for Danish speakers in 2026? With six million native speakers, Danish serves as a stress test for AI localization in smaller languages. From chatbots to speech-to-text and text-to-speech systems, this episode dives into the unique challenges Danish poses, from its complex phonology to healthcare applications. Discover why even a high-resourced language like Danish struggles...
The Cognitive Cost of Punctuation 19.04.2026 28:08
Why does Spanish front-load question marks while English waits until the end? This episode explores how punctuation conventions shape reading efficiency, the hidden cognitive costs of late signals, and what happens when multilingual contexts force us to notice systems we took for granted. Episode #910330 — open it directly at myweirdprompts.com/910330
Blind Ranking AI's Best Podcast Scripts 18.04.2026 35:26
In this unique experiment, 15 large language models—from frontier AI to intentionally flawed ones—were given the same seven controversial prompts to craft podcast dialogues. Hosts Corn and Herman react to the results blind, ranking the models based on their writing quality, factual accuracy, and creativity. From sharp legal debates on Kosovo to witty takes on pronoun norms, discover which models d...
When AI Forecasts Collide: Geopol Model Divergence 18.04.2026 24:51
What happens when five AI models from different training lineages forecast the same fast-moving geopolitical crisis? In this experiment, dubbed the Geopol Forecast Council, models like Claude Sonnet 4.6, Google’s Gemini 3 Flash Preview, and China’s GLM 5.1 independently analyzed the Iran-Israel-US conflict across three time horizons. The results? Both convergence and sharp divergence, offering a r...
Inside Frontier LLM Training: Stages, Costs, and Checkpoints 18.04.2026 26:07
What does it really take to train a frontier large language model? This episode breaks down the multi-stage process, from foundational pretraining to supervised fine-tuning and RLHF. Learn why checkpoints are the backbone of cost efficiency, how labs manage catastrophic forgetting, and why post-training is orders of magnitude cheaper than pretraining. We explore the mechanics of each stage, the st...
Can LLM Councils Truly Capture Diverse Worldviews? 18.04.2026 19:59
Designing an LLM council to maximize diverse perspectives sounds straightforward, but the reality is far more complex. This episode dives deep into whether training corpus diversity translates into worldview diversity after alignment processes like RLHF. We examine models like DeepSeek, Mistral, Falcon, and Jamba, asking if their unique cultural and linguistic training survives the alignment proce...
Ceasefire or Chess Move? Decoding the US-Iran-Israel Triangle 18.04.2026 34:08
The recent ceasefire between Israel and Lebanon has been hailed as a step toward peace, but is there more to the story? This episode unpacks a provocative theory: that the ceasefire is a strategic ruse, shifting responsibility from the US to Israel while Iran and the US quietly engineer an offramp. From the performative conflict in the Strait of Hormuz to Lebanon’s role as a liability shield, we e...
Walking to Jerusalem: The Ancient Pilgrimage Experience 18.04.2026 47:32
Traveling to Jerusalem during the Second Temple period was no small feat. Pilgrims from across the ancient world—Babylon, Alexandria, Rome, and beyond—undertook grueling journeys to reach the Temple, often walking for weeks or months. This episode delves into the physical, social, and spiritual dimensions of ancient pilgrimage, uncovering how the Temple’s design, the structured routes, and the com...
The Serverless Paradox: Why TTS Eats Your Budget 18.04.2026 21:22
Most podcast producers optimize the wrong part of their pipeline. This episode reveals why TTS costs dominate, how batch processing can backfire, and the counterintuitive strategy for cutting infrastructure bills without sacrificing quality. Episode #327298 — open it directly at myweirdprompts.com/327298
How Airlines Build (and Lose) New Flight Routes 18.04.2026 49:31
When an airline announces a new direct route, what’s really happening behind the scenes? This episode explores the complex machinery of air travel, from overfly rights and airport slot negotiations to financial modeling and geopolitical risks. Using the short-lived Israel-Ireland route as a case study, we break down the invisible infrastructure that makes—or breaks—a flight. Learn why launching a...
Inside Podcasting's Simple, Powerful Infrastructure 18.04.2026 1:00:49
Podcasting’s infrastructure is older and simpler than most people realize, yet it remains remarkably powerful. This episode dives into the RSS specification, the backbone of podcasting, and how creators can leverage tools like Vercel and Cloudflare R2 to maintain control over their shows. Learn about the technical decisions behind building a custom podcast feed, the challenges of analytics without...
Why Documentaries Beat White Papers 18.04.2026 26:38
Herman and Corn explain why the best recent documentaries don't just inform—they compress context and make you feel geopolitical and tech shifts before you've decided to care. A practical guide to films that leave you with frameworks, not just facts. Episode #384261 — open it directly at myweirdprompts.com/384261
The Open-Source vs. Commercial Tension in Self-Hosted Media 18.04.2026 45:51
Twenty-five million people run their own media servers, but the real story is the friction between community ideals and corporate survival. This episode explores how Plex, Jellyfin, and Emby navigate the tradeoffs between openness, features, and sustainability. Episode #437260 — open it directly at myweirdprompts.com/437260
The Regulatory Patchwork Behind Hebrew Announcements 18.04.2026 35:36
Why do non-Israeli airlines always have a Hebrew-speaking crew member on flights to Tel Aviv? This episode unpacks the surprising web of international aviation rules, national authority requirements, and airline logistics that produce a consistent practice without a single global mandate. Episode #519927 — open it directly at myweirdprompts.com/519927
How to Scrape Geo-Restricted Israeli Sites with MCP Tools 18.04.2026 29:52
Israeli websites often deploy advanced bot-protection like Cloudflare Turnstile and PerimeterX, making data scraping a challenge. This episode explores practical solutions using MCP tools, residential IPs, and tunneling setups like Tailscale or Cloudflare tunnels. Discover the tradeoffs between Playwright, Firecrawl, and vision-based approaches, and learn why a home workstation architecture might...
When Fleet Size Changes Everything 18.04.2026 35:56
Why does a 50-plane airline operate fundamentally differently from a 500-plane one? This episode explores the qualitative shifts in maintenance, logistics, and strategy that happen at fleet size thresholds, using El Al's Dreamliner expansion as a case study. Episode #967056 — open it directly at myweirdprompts.com/967056
Why Taiwan's Automation Strategy Leaves the West Behind 18.04.2026 23:03
Asus hit 85% automation in its motherboard lines while Western competitors lag at 40-50%. This isn't a technology gap—it's a strategic and cultural one. We explore why Taiwan's posture toward manufacturing automation is fundamentally different, and what the West can learn from it. Episode #866549 — open it directly at myweirdprompts.com/866549
The Side Sleeper’s Edge: Why Most of Us Sleep Curled Up 18.04.2026 23:34
Why do most people sleep on their side, and is it actually better for you? This episode dives into the surprising physiological advantages of side sleeping, from airway optimization to brain waste clearance. We’ll explore the data behind sleep quality, why back sleeping isn’t the universal solution it’s often thought to be, and why side sleepers struggle to switch positions. Whether you’re a side...
Similar podcasts
Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.