x

75% of people globally want unbiased news (Pew Research) but are concerned they're not getting that.

Get balanced coverage with Symby. See how stories are covered across diverse outlets and spot when one side covers a story the other skips.

PODCAST

My Weird Prompts

Daniel Rosehill

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

technology

All Episodes

23:53

How Recommendation Engines Really Work

My Weird Prompts ·

2026/04/21

Ever wonder how Netflix seems to *know* your next binge-worthy show? Behind every "Recommended for You" row is a staggeringly complex AI pipeline—candidate generation, ranking, reranking, and a feature store stitching it all together. This episode breaks down how modern recommendation engines blend battle-tested techniques (like matrix factorization and gradient-boosted trees) with cutting-edge AI (embeddings, two-tower models, and even LLMs). We’ll explore why these systems use cascading stages

en-us

22:37

Iran's Contradictory Signals: Strategy or Chaos?

My Weird Prompts ·

2026/04/21

Iran’s recent moves—simultaneously pushing naval confrontations in the Strait of Hormuz and floating nuclear halt proposals—have left analysts puzzled. Is this deliberate strategy or internal chaos? This episode dives into Iran’s use of contradictory signals as a tool to create ambiguity, drawing parallels to Cold War tactics like Soviet "reflexive control." We explore how this approach grinds down adversaries’ decision-making processes, the challenges it poses for intelligence tradecraft, and the

en-us

38:11

Why LLMs Forget the Middle of Long Conversations

My Weird Prompts ·

2026/04/21

Ever noticed how large language models seem to lose track of things in the middle of long conversations? This episode dives into the science behind this phenomenon, exploring transformer attention mechanisms, positional encodings, and attention dilution. We also discuss practical engineering solutions, like Claude Code’s periodic reminders, and unpack research findings from Stanford’s "Lost in the Middle" paper. Whether you’re a developer or just curious about AI, this episode sheds light on a

en-us

29:50

Building a Custom Home Alarm Panel with ESP32

My Weird Prompts ·

2026/04/21

Ever wished for a physical button to arm your home alarm system? This episode dives into building a custom alarm panel from scratch, designed to integrate seamlessly with Home Assistant. Learn why tactile buttons beat phone apps in moments of stress, how Zigbee sensors create a robust home alarm network, and why ESP32 is the perfect microcontroller for this project. We’ll walk through component choices, from Omron switches to diffused LEDs, and explore the benefits of local control over

en-us

41:01

Building a Dual-Timezone Clock with ESP32

My Weird Prompts ·

2026/04/21

Ever wondered how to build a dual-timezone clock that displays both local time and Zulu time? This episode dives into the nuts and bolts of creating a precise, USB-C-powered clock using an ESP32 microcontroller. From selecting the right components—like LCD displays, an RTC module, and a CR2032 battery backup—to assembling the final product, we cover every step. Discover why precise timekeeping is critical for military operations and how this project offers a clean introduction to ESP32 development

en-us

32:59

Embedded Systems: From Breadboards to Pacemakers

My Weird Prompts ·

2026/04/21

Embedded computing spans a vast spectrum, from the ESP32 on a breadboard to the silicon inside a pacemaker. This episode breaks down the four key categories of embedded systems—MCUs, PLCs, SoCs, and FPGAs—and explores how their design philosophies diverge. Why does a pacemaker use a Cortex-M0 instead of a more powerful chip? How do industrial PLCs survive extreme environments? And what makes FPGAs indispensable for real-time signal processing? The answers reveal why "small computer" doesn't begin to

en-us

45:43

Iran's ICBM Claim vs Anti-Tamper Tech Reality

My Weird Prompts ·

2026/04/21

Iranian state media asserts the capture of American Jericho ICBMs—except the Jericho is an Israeli missile, revealing either ignorance or disinformation. But behind the propaganda lies a real technical question: How do advanced militaries protect sensitive hardware when it falls into adversary hands? From cryptographic zeroization to tamper-proof meshes and geofencing triggers, we break down the layered defenses designed to make reverse-engineering costly and slow. Plus, why Iran’s drone recovery

en-us

32:38

Claude Code Meets Linux Logs: Proactive System...

My Weird Prompts ·

2026/04/21

Linux system administrators have long relied on logs to diagnose issues, but manually parsing through journalctl, syslog, and dmesg can be tedious and error-prone. Enter Claude Code, an AI-powered terminal assistant that automates log analysis, identifies anomalies, and proactively flags potential system failures. This episode explores how Claude Code integrates with Linux’s logging ecosystem—journald, syslog, /var/log, and more—to streamline system administration. Discover how AI can catch errors

en-us

28:48

EU-Israel Agreement: What Suspension Would Really Mean

My Weird Prompts ·

2026/04/21

Spain, Ireland, and Slovenia are pushing to suspend the EU-Israel Association Agreement, citing breaches of human rights clauses. But what would suspension actually mean in practical terms? This episode dives into the three pillars of the agreement—trade, political dialogue, and research cooperation—and explains the legal and political hurdles involved. We also unpack the qualified majority voting system, Germany's pivotal role, and the potential consequences of suspension for both parties. Whether

en-us

19:36

Claude Code As System OS Doctor — Pushing The Limits

My Weird Prompts ·

2026/04/21

Claude Code, marketed as a developer’s coding assistant, has quietly become a favorite among system administrators for managing machines and infrastructure. But its design assumes a developer working within a project repository, creating friction for sysadmin workflows. This episode explores why Claude Code works so well for sysadmins, the limitations of its sandboxing model, and whether its architecture could evolve to better serve this unintended user base. Dive into the nuances of permissions,

en-us

23:03

ESP32 vs Raspberry Pi: The Microcontroller Mindshift

My Weird Prompts ·

2026/04/20

Most developers think of microcontrollers like the ESP32 as just "smaller computers," but the reality is far more interesting. Unlike Raspberry Pis or other single-board computers (SBCs), microcontrollers operate on a fundamentally different tier—running real-time operating systems (RTOS) or even bare-metal firmware, with no traditional OS in sight. This episode dives into the architectural distinctions: deterministic scheduling, memory models, boot times, and power efficiency that make

en-us

21:31

AI Model Spotlight: ** Phi (umbrella brand);...

My Weird Prompts ·

2026/04/20

Dive into Microsoft AI's Phi family of small language models in this episode. Discover how these models prioritize data quality over size, delivering competitive performance with minimal parameters. Learn about their evolution from Phi-1’s coding capabilities to Phi-4’s multimodal features, their edge deployment philosophy, and the benchmarks that validate their efficiency. Whether you're curious about on-device AI or the trade-offs of smaller models, this episode breaks down what makes Phi a

en-us

23:04

How Fast Apply Models Revolutionize AI Code Edits

My Weird Prompts ·

2026/04/20

Frontier AI models like GPT-5 and Claude excel at reasoning but struggle with the mechanical task of merging code edits efficiently. Enter fast apply models — specialized tools designed to stitch AI-suggested changes into source files at blazing speeds of ten thousand tokens per second. This episode dives into the two-model pipeline architecture, why it’s essential, and how tools like Relace Apply 3 are transforming developer workflows. Learn why specialization persists even as frontier models

en-us

20:24

AI Model Spotlight: ** Cogito v2.1 671B

My Weird Prompts ·

2026/04/20

Dive into the inner workings of Cogito v2.1 671B, the latest open-weight large language model from Deep Cogito. Built on DeepSeek-V3-Base, this model introduces process supervision to optimize reasoning chains, reducing token usage by 60% compared to peers. Explore its Mixture of Experts architecture, benchmark performance, and competitive pricing across platforms like OpenRouter and RunPod. Whether you're building a coding assistant or tackling complex reasoning tasks, Cogito v2.1 offers a

en-us

18:36

AI Model Spotlight: **UNKNOWN** — page returned HTTP 404

My Weird Prompts ·

2026/04/20

What happens when you try to profile an AI model and the source page doesn’t load? In this episode, we explore Amazon Nova, a model family hosted on AWS Bedrock, despite the lack of primary source material. We discuss what we can infer about Nova’s architecture, pricing, and potential use cases — and where the gaps in our knowledge lie. Whether you’re curious about enterprise AI or just love a good mystery, this episode offers a candid look at the challenges of analyzing emerging tech.

en-us

20:14

AI Model Spotlight: ** Palmyra X5

My Weird Prompts ·

2026/04/20

Dive into Palmyra X5, Writer’s enterprise-focused AI model, built for long-context retrieval and agentic workflows. With a million-token context window, multi-turn function calls, and native orchestration, Palmyra X5 targets regulated industries like healthcare and finance. This episode unpacks its architecture, benchmarks, pricing, and practical applications, offering insights for teams evaluating AI solutions.

en-us

23:45

Object Detection APIs: Choosing the Right Tool for...

My Weird Prompts ·

2026/04/20

Object detection APIs are essential for tasks like retail inventory management and automated annotation pipelines. But how do you choose the right tool for your workflow? This episode dives into the practical differences between general-purpose vision models like Gemini and dedicated tools like AWS Rekognition, Google Vision API, and YOLO. We explore structured output reliability, precision, cost implications, and the hidden engineering challenges that can make or break your system. Whether you’re

en-us

21:03

AI Model Spotlight: ** Aion-2.0

My Weird Prompts ·

2026/04/20

Aion-2.0, a roleplay-focused AI model, comes from an unlikely source: AionLabs, an Israel-based venture studio specializing in AI-driven drug discovery. Built on DeepSeek V3.2, it boasts a 131k-token context window and unique reasoning token visibility—but its benchmark claims lack independent verification. Why would a pharma-adjacent lab release a storytelling model? We break down its architecture, pricing (with surprising cache efficiency), and why niche platforms like Janitor AI and SillyTavern

en-us

21:35

AI Model Spotlight: ** NVIDIA Nemotron 3 Super

My Weird Prompts ·

2026/04/20

NVIDIA’s Nemotron 3 Super marks a bold step into the AI model arena, blending Mamba, Transformers, and a unique Latent MoE design for unparalleled efficiency. This episode breaks down its architecture, benchmarks, and practical applications, from agentic behavior to long-context reasoning. Learn how NVIDIA’s hybrid approach delivers four times better memory efficiency and three times faster inference, while exploring the model’s strengths and limitations in real-world workloads. Whether you’re

en-us

20:39

AI Model Spotlight: ** Trinity Large Thinking

My Weird Prompts ·

2026/04/20

Arcee AI’s Trinity Large Thinking is making waves in the AI world, and it’s not just because of its impressive benchmarks. Built by a team of just 30 people, this reasoning-optimized model combines sparse Mixture of Experts architecture with innovative efficiency techniques to deliver high performance at a low cost. Dive into the details of its training, architecture, and unique reasoning capabilities, and learn why it’s a standout choice for agentic workloads. We’ll also explore its benchmark

en-us

2,236 results

Similar Podcasts

Lead With AI

Tamara Nall

The Restaurant Technology Guys Podcast brought to you by Custom Business Solutions

Restaurant Technology Guys Podcast,...

Tech Updates

Meir Tsvi

DGTL Voices with Ed Marx

Marx Advisory

The Automated Daily

TrendTeller