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

Technology

Podcast website

myweirdprompts.com

Latest episode

Nov 21, 2026

Where to listen?

Podcasts in the app Replaio Radio Coming soon

Podcasts are coming to the app soon. Install now and be the first to see a whole new take on podcasts

Get it on Google Play Install for free Android 10M+ downloads · 4.7+ rating iOS soon

Episodes

Why You Can't Give an AI the Database Password 15.03.2026

Why can't you just hand an AI agent the keys to your Postgres database? This episode explores the fundamental reasons APIs exist—security, stability, and semantics—and why direct database access is a dangerous siren song for autonomous agents. Episode #426460 — open it directly at myweirdprompts.com/426460

Beyond the Vibes: Mastering Structured AI Outputs 15.03.2026

Tired of LLMs adding conversational filler to your data? This episode explores the technical shift from prompt-based formatting to API-level strict enforcement. We dive into the mechanics of constrained decoding, the evolution of JSON Schema standards, and why libraries like Pydantic are essential for modern AI development. Discover how to use semantic field names and property ordering to improve...

The Architectural Divide Between Batch and Live Speech 15.03.2026

Why does the same AI model feel like a genius in batch transcription but a toddler in real-time voice typing? This episode unpacks the core difference: bidirectional context versus blindfolded guessing, and what it will take to make dictation seamless. Episode #391991 — open it directly at myweirdprompts.com/391991

The Missing Ring Zero: Why LLMs Can't Keep Secrets 15.03.2026

Why can't we just tell an AI to keep its instructions secret? This episode explores the fundamental architectural flaw—the lack of hardware-enforced privilege separation—that makes system prompt leakage inevitable, and what developers can do about it. Episode #196360 — open it directly at myweirdprompts.com/196360

AI Wearables: Local Sovereignty vs. The Subscription Trap 15.03.2026

As AI wearables like the Plaud NotePin and Omi pendant flood the market, users face a critical choice between polished, subscription-heavy ecosystems and raw, open-source hardware that prioritizes data sovereignty. This episode dives deep into the technical architecture of these "remote ears," explaining why high-quality transcription usually requires the cloud and how the latest breakthroughs in...

The Vector DB Hangover: Scaling Without Going Broke 15.03.2026

The "gold rush" of vector databases has ended, replaced by a cold reality of high monthly bills and resource constraints. In this episode, we dive into the true cost of vector storage in 2026, comparing the "RAM tax" of high-performance engines like Qdrant against the cost-saving "mmap" strategies that make $20 servers viable for million-vector indexes. We explore the architectural challenges of s...

Is Your JSON Store Just a Postgres Feature Now? 15.03.2026

Fifteen years after the NoSQL revolution promised to kill the relational database, SQL remains the undisputed industry standard. This episode explores the technical and business reasons why "schema-on-read" flexibility often led to operational debt, and how PostgreSQL eventually neutralized the NoSQL threat by adopting its best features. We also dive into the modern database landscape, discussing...

The Postgres Vector Revolution: Killing the Sprawl 15.03.2026

The rise of AI has sparked a massive gold rush for dedicated vector databases like Pinecone and Weaviate, but the answer to your infrastructure woes might already be sitting in your tech stack. In this episode, we dive into the fascinating history of PostgreSQL and how a design decision made in 1986 paved the way for the modern AI revolution. We explore the "pgvector" extension, comparing its perf...

Escaping JOIN Hell: The SQL Developer’s Guide to Neo4j 15.03.2026

Are your SQL queries buckling under the weight of complex relationships and fifteen-deep JOINs? In this episode, we explore the fundamental shift from relational tables to Neo4j’s graph model, breaking down why "index-free adjacency" is a total game-changer for multi-hop traversals and systemic connections. We move past the hype to examine the practical realities of "relationship intelligence" in...

Why Your AI Is Programmed to Disobey You 15.03.2026

Behind every AI chat box lies a hidden "system prompt"—a complex set of meta-instructions that define the model’s personality, safety guardrails, and boundaries before you even type a word. This episode explores the technical and ethical tension between user intent and vendor control, pulling back the curtain on the "invisible hand" that guides modern LLMs. We dive into the mechanics of instructio...

The Agent-First Shift: Ending the Dual-Track API Tax 15.03.2026

Are you tired of building every feature twice—once for humans and once for AI agents? This episode dives into the "dual-track problem" where developers are currently stuck maintaining separate REST APIs and Model Context Protocol (MCP) definitions, leading to a massive 20% overhead in development velocity. We explore the transition from API-first to agent-first architectures, the role of Google’s...

Beyond Buttons: Is the Admin Dashboard Dead? 15.03.2026

For decades, graphical user interfaces have been the only way for humans to manage complex digital systems, but that era is coming to a close. This episode explores the revolutionary shift toward the Model Context Protocol (MCP), a framework that allows AI agents to bypass visual dashboards and interact directly with system backends. We discuss how "headless admin" setups are making traditional in...

The Hidden Technical Debt in Your Wrist 15.03.2026

Why do we still use 1980s mouse designs for 10-hour workdays? This episode examines the physiological cost of forearm pronation, the case for vertical mice, and why your input stack might be your biggest health liability. Episode #213307 — open it directly at myweirdprompts.com/213307

The Hidden Math of Readability 15.03.2026

Lexile, Flesch-Kincaid, and the Goldilocks problem of cognitive load: how algorithms measure the perfect difficulty for young readers, and why syllable counting misses the story. Episode #293034 — open it directly at myweirdprompts.com/293034

System Update: Navigating the 9-Month Growth Spike 15.03.2026

At nine months, infants undergo a massive "system update," transitioning from passive observers to active explorers of their environment. This episode dives into the cognitive and physical shifts that define this volatile period, from the emergence of independent mobility to the complex development of object permanence. We explore why this stage feels so chaotic for parents and why the infant brai...

Rethinking Mastery: Beyond the 10,000 Hour Rule 15.03.2026

For decades, the 10,000-hour rule has been the gold standard for achieving mastery, but in the rapidly shifting technological landscape of 2026, this metric is fundamentally broken. This episode dives into why software engineering is an "open system" where skills decay faster than they can be acquired through repetition. We explore the critical distinction between deliberate practice and "muscle m...

The Algorithmic Adversary: Inside the IRGC’s AI Strategy 15.03.2026

In this deep dive, we move beyond the kinetic "bang" of traditional warfare to examine the rise of the algorithmic adversary. The Islamic Revolutionary Guard Corps (IRGC) is no longer just a regional spoiler; it has evolved into a sophisticated technological actor using artificial intelligence as the ultimate asymmetric force multiplier. We explore the mechanisms of "Information Attrition," where...

The End of Negotiation: A New Kinetic Order 15.03.2026

Why the era of diplomatic off-ramps has collapsed and what the 'Kinetic Core' of US-Israel operations means for global security, as regional partners like Azerbaijan and the UAE embrace hard-power realism. Episode #129939 — open it directly at myweirdprompts.com/129939

Can Your Air Defense Handle the Math of a 400-Missile Salvo? 15.03.2026

The rules of engagement have crossed a tactical Rubicon, moving from symbolic signaling to high-volume saturation campaigns. This episode analyzes the "March 12th event" to reveal how solid-fuel systems and maneuverable reentry vehicles are rendering traditional air defenses obsolete. Explore why military planners are abandoning the dream of a "perfect shield" in favor of a grim new reality: strat...

Fraying the Ring of Fire: The Collapse of Iranian Proxies 15.03.2026

The Middle East is witnessing a seismic shift as the Iranian "Ring of Fire" begins to fray under intense coalition pressure. This episode analyzes the strategic dismantling of IRGC logistics nodes and what happens when a global proxy network loses its patron. Explore how this collapse is forcing a new security-first reality that is reshaping regional alliances and the future of the Abraham Accords...

When Biology Becomes a Garage Hobby 15.03.2026

AlphaFold 3 has turned the protein folding problem from a grand challenge into a laptop app. This episode explores what happens when the blueprint of life becomes accessible to anyone—from dog owners designing mRNA vaccines in garages to the dual-use risks of democratized biology. Episode #602551 — open it directly at myweirdprompts.com/602551

Swarm-as-a-Service: How Cheap Drones Broke Air Defense 15.03.2026

Modern air defense is facing a "DDoS attack" in physical space as low-cost Iranian drones like the Shahed-136 overwhelm sophisticated radar systems. By utilizing off-the-shelf components and flying at speeds that mimic biological clutter, these "mopeds with explosives" force defenders into a lopsided cost-exchange ratio that is redefining the economics of warfare. This episode breaks down the tech...

The Overseas Front: Iran’s Global Campaign of Unrest 14.03.2026

Recent intelligence briefings from early 2026 suggest a paradigm shift in the nature of global antisemitic violence, moving away from spontaneous domestic unrest toward a highly coordinated strategic operation directed by Tehran. This episode explores the emergence of the "overseas front," a doctrine of asymmetric domestic disruption where the Iranian Revolutionary Guard Corps (IRGC) leverages Wes...

The Myth of Lawless Seas 14.03.2026

Dreaming of escaping to international waters to avoid taxes and rules? The reality is a dense web of treaties and jurisdictions that govern every nautical mile. This episode unpacks UNCLOS and why the ocean is far from a lawless frontier. Episode #820470 — open it directly at myweirdprompts.com/820470

The Blind Spot: When Radar Gaps Make Every Missile a Nuclear Threat 14.03.2026

With critical radar arrays destroyed in Qatar and Jordan, the U.S. and Israel now face a terrifying scenario: minutes to determine if an incoming missile carries conventional explosives or a nuclear warhead. This episode examines the physics, psychology, and policy failures behind pre-launch ambiguity. Episode #187437 — open it directly at myweirdprompts.com/187437

Listen to the My Weird Prompts podcast in Replaio

Radio and podcasts in one app - free, with no sign-up. Install today and do not miss the launch

Get it on Google Play

Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.