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
Non-Coders Are Hijacking the Terminal 04.04.2026 23:23
The command line is no longer just for developers. Researchers, writers, and analysts are turning terminal-based AI agents into powerful productivity workspaces—without writing a single line of code. From managing equity research to organizing personal therapy notes, these "non-coders" are redefining what these tools can do. We explore the three pillars making this possible: repo-as-workspace, per...
Pixels vs Protocols: The Computer Use Showdown 04.04.2026 27:44
The podcast explores the architectural tension between visual "Computer Use" agents—like Anthropic's demo—and API-first automation. Hosts analyze whether visual agents are a high-latency bridge to a protocol-driven world or a necessary tool for legacy systems. They discuss cost implications, reliability issues, and the potential for visual interaction to become just another capability rather than...
Saving AI Knowledge Beyond the Chat Window 04.04.2026 24:44
Every day, companies lose massive amounts of institutional intelligence because AI chat outputs are treated as disposable. In this episode, we explore the "ephemeral context trap" — the gap between brilliant AI conversations and permanent knowledge bases. We discuss why current tools fail to capture the "trail of thought," and outline a five-step pipeline (Capture, Sanitize, Extract, Categorize, H...
Building Better AI Memory Systems 04.04.2026 22:21
What happens to your AI's brilliant answers after you see them? In this episode, we explore the "leaky bucket" problem of AI output storage. We discuss why treating AI conversations as ephemeral is a corporate nightmare, and dive into the tools trying to give these models a long-term memory. From LangSmith and Langfuse to "Reverse RAG" and projects like Mem zero and Letta, we uncover how to turn a...
The Plumbing of AI Safety: Guardrails, Not Vibes 04.04.2026 23:49
We move past vague ethics to the literal plumbing of AI safety. This episode explores the specific libraries, proxy layers, and architectural decisions that act as the new enterprise firewall for LLMs. We dissect the tension between latency and security, comparing "sandwich" guardrails with token-level steering, and break down the open-source versus commercial landscapes—from NVIDIA NeMo and Guard...
Needle-in-a-Haystack Testing for LLMs 04.04.2026 20:49
We have massive AI models that claim to be "world-class intelligent," yet they often fail at basic tasks like finding a specific fact in a long document. This episode explores the disconnect between benchmark scores and real-world performance, diving into EvalScope, an open-source toolkit designed to stress-test long-context retrieval and agentic capabilities. We discuss the "lost in the middle" p...
AI Grading AI: The Snake Eating Its Tail 04.04.2026 22:15
The industry is scaling faster than humans can review, so we’ve turned to LLM-as-a-Judge to grade model outputs. But this creates a hall of mirrors: AI grading AI, often with a preference for verbosity and its own style. We explore the mechanics of single-point, pairwise, and reference-based scoring, and the hidden biases—like position and self-enhancement—that threaten to create a monoculture of...
How Do You Measure an LLM's "Soul"? 04.04.2026 22:36
We all know how to test if an LLM solves a math problem, but how do you measure if it has the right "soul"? This episode tackles the messy world of qualitative AI evaluation. We explore why binary benchmarks fail for real-world tasks like medical summaries or creative writing, and dive into techniques like LLM-as-a-Judge, G-Eval, and counterfactual testing to map a model's hidden worldview. Learn...
Beyond Vibes: The Hard Science of LLM Evaluation 04.04.2026 23:35
How do you know if your LLM is actually working? This episode moves past gut checks to explore the real metrics—coherence, hallucination, instruction-following—and why even your hardware choice can skew results. Episode #786706 — open it directly at myweirdprompts.com/786706
The AI Control Plane Is Here (But Is It Safe?) 04.04.2026 24:54
As AI agents move from prototypes to production, teams face a fragmented mess of inference gateways, MCP servers, and observability tools that don’t talk to each other. This episode explores the rise of the "AI Control Plane"—a unified infrastructure layer that promises a single pane of glass for routing models, managing tools, and tracking costs. We dig into how these systems handle security, con...
The Velocity Paradox: Why Faster Code Means Slower Ships 04.04.2026 25:58
When AI agents can execute code instantly, the cost of a wrong direction skyrockets. We explore the "Velocity Paradox" in modern development, where the ease of building creates new psychological traps like scope creep, architectural debt, and the loss of the "gut check." Learn how to manufacture friction through Idea Backlogs, Triage, and Spec-Driven Development to ensure your speed actually leads...
Brainstorming a Stable-by-Design Smart Home 04.04.2026 21:12
Home Assistant's fragility stems from its monolithic architecture. Could sandboxed 'Integration Pods' using WebAssembly or containers give us granular control without the late-night YAML debugging? Episode #359440 — open it directly at myweirdprompts.com/359440
Stop Writing "It Feels Slow" Tickets 04.04.2026 20:30
We’ve all seen it: a ticket that just says "The app feels slow." But what actually makes a bug report useful? This episode dives into the high art of bug reporting, from the "Golden Trio" of information to the "ping-pong" effect that kills productivity. We explore the modern landscape of issue tracking tools—from the enterprise heavyweight Jira to the developer-loved Linear—and look at the new wav...
Why Intelligence Agencies Slice the World into Desks 04.04.2026 23:12
Every superpower sees the world through a bureaucratic map of "desks"—but these divisions are often Cold War ghosts that create dangerous blind spots. This episode explores how the CIA, State Department, and Pentagon draw different borders, why Egypt sits in a military turf war, and how the "seam" between Afghanistan and Pakistan caused chaos during the 2021 withdrawal. You’ll learn why desk offic...
Why Anti-Zionist Jews Live in Jerusalem 04.04.2026 21:55
Explore the theological paradox of religious Jews who oppose the State of Israel but choose to live in Jerusalem. This episode dives into the Talmudic "Three Oaths," the history of the Satmar and Neturei Karta movements, and the distinction between the holy Land and the secular State. Learn why these communities refuse government funding, avoid the draft, and navigate a life of ideological frictio...
Why Your Brain Lies About Where the Bomb Is 04.04.2026 26:17
When a flash in the sky looks overhead but the explosion happened miles away, your survival instincts betray you. This episode explores the physics of sound and light latency in conflict zones, and why your primate brain can't handle high-altitude warfare. Episode #889642 — open it directly at myweirdprompts.com/889642
The Long Peace Is Over (Or Is It?) 04.04.2026 18:27
Is humanity actually getting safer, or are we just in a lucky lull before catastrophe? We dig into the data on the "Long Peace" since 1945, examining the three key suppressors of war—nuclear weapons, economic interdependence, and international institutions—and why that peace might be fraying at the edges. From the statistical nadir of 2010 to the rising conflict counts of 2026, we explore the deba...
Why Leaders Broadcast Victory While Citizens Hear Sirens 04.04.2026 33:14
Why do leaders broadcast polished statements while citizens face a different reality? This episode explores the "hermetic shield" of modern communication, comparing FDR's fireside chats to today's curated feeds. We examine how the gap between official narratives and live data erodes public trust and what it means for leadership in 2026. Episode #893372 — open it directly at myweirdprompts.com/8933...
The Human Curriculum Machine 04.04.2026 28:11
We worry about AI bias in education, but the human system is already compromised. This episode deconstructs the massive, clanking machine that decides what kids learn before they even start school. Discover the "Texas Effect," why nearly 80% of teachers ignore official textbooks, and how budget deals override pedagogy. Episode #819182 — open it directly at myweirdprompts.com/819182
Why Can't AI Admit When It's Guessing? 04.04.2026 30:31
As AI research agents scan thousands of documents, they increasingly auto-flag their own uncertain claims. But how reliable is this "self-awareness"? We explore the mechanics of confidence scoring in LLMs, from simple self-reports to advanced multi-agent auditing and calibration layers. Discover why a model's certainty often doesn't match its accuracy, and how engineers are building rigorous verif...
Hiding the Kitchen: Why AI Shouldn't Show Its Work 04.04.2026 31:14
Why should AI chatbots hide their messy back-end processes? This episode explores the orchestrator-worker model as a way to protect user attention, debating whether spawning sub-agents is brilliant or wasteful, and what this means for the future of software interaction. Episode #260714 — open it directly at myweirdprompts.com/260714
The Sovereign Compute Shift: Owning vs. Renting AI Iron 04.04.2026 34:14
Nations are moving from renting cloud time to building national AI supercomputers. This episode explores the technical architecture of distributed GPU clusters, why lower-precision math drives AI efficiency, and how Israel's 4,000-GPU system aims to prevent brain drain. Episode #280334 — open it directly at myweirdprompts.com/280334
Why 20 Clean Qubits Beat 1000 Noisy Ones 04.04.2026 25:46
Israel's new 20-qubit quantum computer isn't about competing on scale—it's about sovereignty, precision, and controlling the stack. We explore why a smaller, domestically built machine can outmaneuver larger foreign systems in defense and industry. Episode #256766 — open it directly at myweirdprompts.com/256766
Education’s Robot Problem: Standardization vs. Self-Direction 04.04.2026 23:02
Is the traditional degree becoming obsolete? This episode dives into the tension between standardized education and the rising value of self-directed learning in an AI-driven world. We explore how industries like medicine are blending core competencies with learner autonomy, and why the "Carousel Model" might be the future of higher education. From IBM's "New Collar" initiatives to the mastery tra...
Your Cloud Photos Vanish If You Miss a $5 Bill 04.04.2026 21:44
We explore the hidden fragility of cloud archival storage versus the home NAS approach. Learn about the "retrieval trap" costs, the risk of automated data deletion, and the practical strategies—like Object Lock and the 3-2-1-1 rule—needed to keep your digital memories safe in 2026. Episode #647536 — open it directly at myweirdprompts.com/647536
Similar podcasts
Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.