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 22, 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
How a $15 Radar Reveals the Hidden Economy of Scale 29.12.2025 20:27
How does a millimeter-wave radar sensor end up at your door for fifteen dollars? This episode unpacks the CMOS revolution, Shenzhen's hardware clusters, and the shipping subsidies that make the impossible affordable—and asks what we sacrifice for convenience. Episode #948157 — open it directly at myweirdprompts.com/948157
The Agentic AI Dilemma: Who Holds the Kill Switch? 29.12.2025 21:09
In this episode of My Weird Prompts, Herman and Corn dive into the complex world of agentic AI and the critical necessity of human oversight. They discuss the shift from simple chatbots to autonomous agents managing power plants and medical diagnostics, exploring frameworks like "human-on-the-loop" and "formal verification." From the psychological trap of automation bias to the unsettling reversal...
The Comeback of RNNs: Why Old Tech Still Matters 29.12.2025 21:21
Herman and Corn bust the myth that deep learning is the only AI game in town, exploring why recurrent neural networks are making a surprising return through liquid networks and state-space models. Episode #296620 — open it directly at myweirdprompts.com/296620
Why Your AI Is a Yes-Man 29.12.2025 26:33
Why do AI assistants always agree with you? This episode unpacks how RLHF's reward models inadvertently train sycophancy, turning chatbots into polite pushovers—and what that means for trust. Episode #012814 — open it directly at myweirdprompts.com/012814
Silencing the Siren: Real-Time AI Noise Reduction 29.12.2025 22:04
In this episode, Herman and Corn dive into the fascinating world of deep neural networks and their role in cleaning up messy audio on mobile devices. From the challenges of "non-stationary" noises like sirens to the engineering trade-offs of running AI on mobile NPUs, they explore how 2025's hardware is changing the way we communicate. They discuss the shift from cloud-based processing to edge com...
The 1% Rule: Mastering Kaizen for Lasting Improvement 29.12.2025 19:18
Are you feeling the end-of-year pressure to "move fast and break things"? In this episode of My Weird Prompts, Herman and Corn dive deep into the Japanese philosophy of Kaizen to help listener Daniel overcome the urge to rush through projects. They break down the history of the Toyota Production System, the math behind the "1% rule," and practical frameworks like the PDCA cycle and 5S. Whether you...
AI in 2025: Is Small the New Big? 28.12.2025 20:59
In this episode of My Weird Prompts, brothers Herman and Corn Poppleberry dive into a provocative thought experiment: if cloud inference costs were identical, would there ever be a reason to choose a small model over a trillion-parameter giant? Moving beyond the "bigger is better" hype of previous years, the duo explores the physical realities of latency, the hidden costs of model verbosity, and t...
From Keywords to Vectors: How AI Decodes Meaning 28.12.2025 18:30
Ever wonder why you can search for "banana bread" with typos and get results, but your own computer fails to find a document if you miss one letter? In this episode of My Weird Prompts, Herman and Corn break down the shift from literal keyword matching to semantic understanding. They explore the fascinating history of "word math," from the linguistic theories of the 1950s to the revolutionary Tran...
How AI Deciphers Your Typo-Ridden Prompts 28.12.2025 25:29
Why does an AI understand 'I rly want a pizz' perfectly? This episode unpacks tokenization and probability to reveal how models see through typos and bad grammar—and when that ambiguity can backfire. Episode #787187 — open it directly at myweirdprompts.com/787187
The Right to Breathe: Tobacco Policy and the Enforcement Gap 28.12.2025 23:43
In this episode, Herman and Corn dive into the suffocating reality of second-hand smoke and the widening gap between global tobacco legislation and real-world enforcement. Inspired by a listener's struggle with asthma in Jerusalem, the discussion moves from the chemical dangers of sidestream smoke to the pioneering bans in Ireland and France. We explore the radical "end-game" strategies of 2025, i...
When Your Body Becomes a Chemistry Set 28.12.2025 24:10
Daniel, a 36-year-old listener, takes five daily medications and wonders if there's a limit. This episode explores the personal anxiety of polypharmacy, the metabolic limits of the liver, and the prescribing cascade—asking how we can manage health without losing ourselves in pills. Episode #925864 — open it directly at myweirdprompts.com/925864
The Hidden Chemistry of Cheap Plastic 28.12.2025 19:02
Why do some plastics reek while others don't? This episode unpacks how manufacturers cut corners with plasticizers and off-gassing, and what that means for your health at home. Episode #519441 — open it directly at myweirdprompts.com/519441
Why Your Smart Bulbs Fail and Airports Don't 28.12.2025 24:56
Why do home smart lights glitch while airport systems run flawlessly? This episode explores the philosophy of industrial automation—from PLCs to Bacnet—and what consumer tech can learn from systems built for permanence. Episode #716331 — open it directly at myweirdprompts.com/716331
Beyond Transformers: Solving the AI Memory Crisis 27.12.2025 21:46
In this episode, Herman and Corn Poppleberry tackle one of the most frustrating hurdles in modern AI engineering: the "stateless" architecture of Large Language Models. They explore why current models require you to resend your entire conversation history with every message, leading to skyrocketing token costs and the "lost in the middle" phenomenon that plagues even the most advanced systems. Fro...
Why Agentic AI Needs More VRAM Than You Think 27.12.2025 21:13
When your AI model tries to debug and test code autonomously, it demands massive memory. Herman and Corn explore why 12GB of VRAM isn't enough, and how to build a local inference server that actually keeps up. Episode #462032 — open it directly at myweirdprompts.com/462032
Beating Context Bloat with Dynamic Dictionaries 27.12.2025 23:29
When a custom dictionary slows your AI to a crawl, how do you keep it hearing niche terms without breaking the bank? This episode unpacks phonetic indexing, dynamic hints, and portable JSON structures as alternatives to fine-tuning. Episode #932370 — open it directly at myweirdprompts.com/932370
Why Your AI Code Assistant Gets Worse 26.12.2025 23:26
Why do AI coding tools feel brilliant at first, then suddenly struggle? This episode unpacks the real reasons behind 'model rot'—from vendor optimization tricks to the psychology of firm prompting—and what developers can do about it. Episode #046174 — open it directly at myweirdprompts.com/046174
The $5.5 Million Breakthrough: DeepSeek’s AI Disruption 26.12.2025 17:41
In this episode of My Weird Prompts, Herman and Corn dive deep into the seismic shift occurring in the artificial intelligence landscape as Eastern models like DeepSeek and Z.ai challenge the status quo. While Western giants like OpenAI and Anthropic spend hundreds of millions on training, DeepSeek has managed to produce world-class performance for a mere $5.5 million. The duo explores the technic...
From Chatbots to Computer-Use Agents 26.12.2025 27:06
How Grace Hopper's vision of natural language computing is finally becoming real through computer-use agents and the Model Context Protocol, turning your AI from a talker into a doer. Episode #904797 — open it directly at myweirdprompts.com/904797
Beyond Math Puzzles: The Truth About AI Benchmarks 26.12.2025 22:25
In this episode of My Weird Prompts, Herman and Corn tackle the growing controversy surrounding artificial intelligence benchmarks. As new models like Claude 4.5 and GLM 4.7 dominate headlines with record-breaking scores, the duo explores whether high performance on math puzzles actually translates to real-world coding productivity. They break down the dangers of data contamination, the rise of "b...
Vibe Coding & The Rise of the AI Orchestrator 25.12.2025 24:15
Are we witnessing the end of the traditional programmer? In this episode of My Weird Prompts, Herman and Corn dive into the world of agentic development and "vibe coding," exploring how tools like Claude Code are shifting the focus from syntax to systems thinking. They discuss how the role of the developer is evolving into that of an "orchestrator," where managing AI agents is more critical than m...
Why Your Brain Might Be Wired for SQL 25.12.2025 23:44
Herman and Corn explore the idea that programming languages fit different brains, not a ladder of difficulty. As AI handles syntax, the future developer is defined by cognitive fit, not coding chops. Episode #572220 — open it directly at myweirdprompts.com/572220
The ADHD Med Maze: Bureaucracy vs. Brain Health 24.12.2025 24:38
In this episode of My Weird Prompts, Corn and Herman Poppleberry tackle the "man-made" crisis of ADHD medication regulation. From the absurdity of counting individual pills in a glove box to the rigid DEA quotas that leave shelves empty, they explore why the system treats patients like suspects rather than people in need of care. Join the duo as they debate the philosophy of access, compare the ri...
AI as a Mirror: Mapping Your Philosophical Identity 24.12.2025 19:28
In this episode of My Weird Prompts, Corn and Herman tackle a fascinating question from their housemate Daniel: Can AI help us label and explore our own personal philosophies? Moving beyond productivity and coding, the duo discusses how Large Language Models act as "high-speed librarians" that bridge the gap between human intuition and academic vocabulary. They dive into current tools like Edubrai...
The Mic That Hears You from Across the Desk 24.12.2025 24:16
Why consumer microphones fail at voice dictation and how professional boundary mics can deliver 99% accuracy without a headset. Herman and Corn explore the audio engineering behind hands-free computing. Episode #508311 — open it directly at myweirdprompts.com/508311
Similar podcasts
Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.