Advertise on podcast: Not Brothers

Everything is “AI slop” now—or at least that is what the internet calls any AI-assisted work it does not like. But is AI actually the problem? In Episode 24 of Not Brothers, Mark Hughes and Ryan Hughes unpack what AI slop really is, why the person piloting the tool still owns the result, and how lazy review—not AI alone—turns fast output into bad work. They share a real example from Oodle’s early agent experiments, when an AI-assisted request led to an unnecessary Python server instead of a lightweight solution. The lesson was not to stop using AI. It was to build better guardrails, encode stronger standards, and create review systems that catch bad decisions before they ship. The conversation also explores why agentic workflows do not fit neatly into old sprint plans, how visual progress can hide weak architecture and poor mobile performance, and why compliance requirements such as GDPR, CCPA, COPPA, WCAG, and CAN-SPAM still demand experienced judgment. For agencies and clients, AI creates a new kind of collaboration. Business leaders can prototype ideas more clearly, teams can explore more directions without burning a week of labor, and experts can turn rough concepts into production-ready work. That does not make expertise less valuable—it changes where the value lives. Mark and Ryan also tackle the economics of AI-enabled services, the “good, fast, and cheap” tradeoff, and why the human side of professional services becomes more important as production accelerates. The takeaway: AI is a force multiplier. It can multiply garbage, or it can multiply great work. Either way, there is still a person piloting the ship. Oodle: https://heyoodle.com/ Chapters are below. Subscribe for direct conversations about technology, AI, business, and the systems behind the work. Chapters00:00 — What counts as AI slop? 01:44 — If you would not put your name on it, it is not finished 03:06 — The rogue Python server and why guardrails matter 04:41 — AI superpowers and “slop grenades” can both be real 05:50 — Defined outcomes, automated review, and responsible shipping 07:42 — When “AI slop” becomes a lazy insult 09:06 — AI did not invent bad process 10:15 — Why old sprint plans break in an agentic workflow 12:33 — Adversarial feedback and the limits of automated code review 14:25 — Fast prototypes can hide production problems 16:05 — Compliance risks an AI agent may never mention 18:04 — In-housing, expertise, and accountability 20:03 — AI as a richer collaboration tool for clients 23:21 — Exploring more ideas without burning the whole budget 25:25 — More value, less effort, and the agency margin question 29:15 — Good, fast, and cheap still has limits 31:08 — Why the human component matters more, not less 32:43 — AI multiplies whatever quality already exists 36:32 — Not everything is AI slop—and kindness still matters

Two weeks, three events, and one clear takeaway: the old way of building software is over. Ryan returns from Rails World, OpenAI DevDay, and an Omarchy meetup in San Francisco with a field report on where AI-assisted development is actually heading. He and Mark compare the conversations happening online with what builders are saying in the room. At Rails World, they unpack the changing role of programmers, why AI does not erase hard-earned technical judgment, and how agents make it easier to choose the right language for the job instead of staying trapped in one stack. Then the conversation moves to OpenAI DevDay: proactive assistants, connected agents, new model performance, and ultra-fast inference. If an agent can produce an hour of work in ten minutes, are you paying more—or buying time back? That leads to the next problem. When generation stops being the bottleneck, human review, browser use, APIs, research speed, infrastructure, and access to compute start taking its place. Finally, Ryan shares what it was like attending his first Omarchy meetup at Cloudflare’s San Francisco office, from veteran users to someone installing Omarchy for the first time at the event. Subscribe for long-form conversations about AI, software, business, operations, and building things in public. Not Brothers: https://heyoodle.com/not-brothers/ Oodle: https://heyoodle.com/ Chapters 00:00 — Ryan’s two-week developer event tour 01:43 — Rails World and the reaction to a changing paradigm 06:11 — “Look at all the things I’m not doing” 11:26 — Choosing the right language for the job 15:31 — Ryan’s biggest Rails World takeaway 18:09 — Why agents suddenly feel more capable 22:23 — Small, specialized models and private AI 27:04 — Inside OpenAI DevDay 28:18 — Proactive assistants and connected agents 32:23 — Faster models and ultra-fast inference 35:10 — Creating at the speed of thought 37:45 — Is buying back time worth the price? 39:29 — When the human becomes the bottleneck 44:56 — Compute demand, hardware prices, and supply 49:02 — Ryan’s first Omarchy meetup 50:46 — Final recap

What happens when a lifelong Mac and Windows user tries Omarchy for the first time? In Episode 22 of Not Brothers, Mark Hughes installs the Try Omarchy experience on his Mac while Ryan Hughes walks him through the operating system, its keyboard-first workflow, and its tiling window manager. Nothing was rehearsed, so the ARM64 installation gaps and setup surprises stay in the episode. Once Omarchy is running, Mark learns how to move between workspaces, rearrange windows, launch apps, pin a window across desktops, and use the searchable keybinding menu when he gets lost. Then the experiment gets more interesting. Mark asks Claude Code to build a custom switcher that combines applications and browser tabs into a Windows-style Alt+Tab experience. Claude inspects Omarchy’s plugin system, builds the plugin, tests it, and updates the live keybinding. Mark gets a working feature without writing the code himself. Ryan also explains T3 Code and T3 Connect, how he gives AI agents their own development machine, why Omarchy treats the operating system as something users should be able to reshape, and how community plugins are expanding what the system can do. Try Omarchy: https://omarchy.org/ Omarchy plugins: https://plugins.omarchy.org/ Learn Omarchy: https://learnomarchy.com/ Oodle: https://heyoodle.com/ Chapters are below. Subscribe for direct conversations about technology, AI, business, and the systems behind the work. Chapters00:00 — A Mac user tries Omarchy 02:33 — Starting the Try Omarchy installation 05:40 — Launching the virtual machine 06:52 — Finishing the Omarchy installer 08:48 — First look at Omarchy and searchable keybindings 10:24 — The Omarchy launcher and installing T3 Code 11:43 — Updating Omarchy before installing apps 13:31 — The ARM64 compatibility gap 14:22 — How Omarchy installs AI tools on demand 15:29 — Signing into Claude Code and 1Password 17:58 — Learning tiling window management 21:02 — Omarchy Quattro’s plugin architecture 22:07 — Asking Claude to build a browser-tab switcher 24:12 — How an agent reverse-engineered a mouse plugin 27:07 — T3 Code, T3 Connect, and remote agent sessions 32:04 — Can the plugin access every browser tab? 35:38 — Workspaces, screen sharing, and a live crash 37:24 — Resuming the Claude Code session 39:50 — The first custom plugin works 40:50 — Why Omarchy is more than Arch and dotfiles 43:41 — Browsing and installing community plugins 46:11 — Combining browser tabs and applications 49:05 — Replacing Omarchy’s default Alt+Tab behavior 49:58 — Making computers fun and personal again 52:41 — How easy is it to try Omarchy? 53:05 — Learn Omarchy and interactive onboarding 57:36 — Custom keybindings and the learning curve 59:53 — Floating, resizing, and pinning windows 1:01:22 — Final verdict on Try Omarchy

Omarchy is moving fast—and AI is changing both what the project can build and what its security team can find. In Episode 21 of Not Brothers, Ryan Hughes and Mark Hughes cover Omarchy’s new website, new infrastructure backing, global meetups, and the people now working full time on the operating system and its open-source foundations. Ryan also breaks down his experience with GPT-6 Astra, Claude Fable 5.1, Meta’s Muse Spark 1.3, and computer-use agents that can navigate a real machine. That power comes with a catch: an agent with access to your computer may use that access in ways you never expected. The conversation moves from frontier models to a more practical question: when should a fast, inexpensive specialist model do the work instead? Ryan shares examples where smaller models matched—or beat—larger models for narrow tasks at a fraction of the cost. Then they get into local AI, HIPAA and data privacy, the rise in reported software vulnerabilities, and why fixing more security issues can be a sign of better visibility rather than worse engineering. Explore Omarchy: https://omarchy.org/ Omarchy news: https://omarchy.org/news/ Omacom Foundation: https://omarchy.org/foundation/ Cua computer-use agents: https://cua.ai/ Chapters are below. Subscribe for more direct conversations about technology, business, AI, and the systems behind the work. 00:00 — Omarchy, AI, and security 00:43 — Omarchy’s new website and a huge week of growth 04:32 — The global Omarchy meetup community 06:06 — GPT-6 Astra vs. Claude Fable 5.1 07:17 — Giving AI agents their own computer 09:59 — The agent that found an API key and kept spending 11:58 — Is this actually AGI? 13:15 — Why agent harnesses and memory matter 15:03 — Using AI to resume complex work 16:37 — Meta Muse Spark 1.3 and a 22-cent site migration 18:57 — Orchestrator models and cheaper executors 20:13 — Why the biggest model is not always the best model 23:06 — Why local AI could break through in 2027 24:30 — Specialist models vs. giant generalists 26:24 — Local models, HIPAA, and data privacy 28:01 — Security is everywhere—and imperfect everywhere 29:35 — A bizarre USB-drive vulnerability 30:50 — Why more security fixes can be a good sign 33:11 — Open-source transparency vs. security through obscurity 35:46 — The coming security boom 36:05 — Closing

No agenda. No single topic. Somehow, still a theme. In Episode 20 of Not Brothers, Mark Hughes and Ryan Hughes start with the latest work around Omarchy, including security tradeoffs, a broader user base, and the Omacom Foundation’s growing investment in desktop Linux. Then the conversation moves into AI-generated music, whether prompting still counts as creativity, and what changes when AI agents can communicate and delegate work to one another. Ryan shares a practical multi-agent security experiment that exposed how easily an authorized agent can become a relay for an unauthorized request. From there, Mark and Ryan look at agent-driven shopping and advertising: how brands may influence AI recommendations, why manipulative source content could become the next black-hat marketing tactic, and whether influencer marketing could eventually shape what models treat as common knowledge. They also get into the FTC and states’ allegations against Amazon’s advertising auctions, Meta’s youth-addiction settlement, corporate penalties that become a cost of doing business, device tracking, cookie banners, and why interruptive advertising still deserves to die. It is a little AI, a little advertising, a little Big Tech, a little conspiracy theory, and exactly enough politics to make sure the episode could not stay untitled forever. Not Brothers is business and tech talk with no nonsense from Mark and Ryan Hughes—Oodle operators, family members, and definitely not brothers. Chapters 00:00 — The episode with no title 01:05 — Omarchy funding, security, and a broader audience 06:27 — AI-generated music and who gets to define art 10:16 — Rogue AI agents and the Hugging Face incident 13:29 — When agents talk to other agents 17:37 — Agent-driven shopping and advertising 19:24 — Can brands manipulate AI recommendations? 21:46 — Influencer marketing as model training data 22:47 — The Black Mirror version of agentic advertising 24:24 — Why interruptive advertising deserves to die 25:52 — Amazon’s alleged ad-auction manipulation 27:59 — Meta, youth addiction, and Big Tech incentives 28:43 — When fines become a cost of doing business 33:21 — Are our devices still listening? 35:20 — Responsible tracking and broken audience cohorts 37:47 — Demographics, geographics, and creepy psychographics 38:33 — Cookie banners and a universal privacy control 40:24 — A spam-reporting model for predatory ads 41:12 — Somehow, they solved advertising

Omarchy began as David Heinemeier Hansson’s opinionated Arch Linux setup. Barely more than a year later, it has grown into a distinct Linux distribution, a fast-moving open-source ecosystem, and the centerpiece of the $10 million Omacom Foundation. In Episode 19 of the Not Brothers Podcast, Ryan Hughes explains how he went from building his own Arch-based setup to contributing directly to Omarchy alongside DHH. He walks through the project’s evolution from configuration files into a complete operating-system experience with its own installer, package repositories, mirrors, stable release channel, plugins, and hardware support. Mark approaches the conversation as a Windows and macOS user who has never lived in Linux. Ryan makes the case for Omarchy through tiling window management, keyboard-driven workflows, fast installation, strong defaults, and the idea of a “malleable computer” that people can shape around the way they actually work. They also dig into Omarchy Quattro, the rapid growth of the plugin ecosystem, and the effort to make AI agents first-class parts of the operating system. That includes crash diagnosis, a built-in agent interface, and automated review processes for a community that submitted more than 1,000 plugins in a week. Ryan shares the hard lessons too: package updates that broke systems overnight, bootloader changes that forced emergency fixes, and the need to balance Arch Linux’s bleeding edge with a stable experience for people trying Linux for the first time. The conversation closes with what this work brings back to Oodle: deeper technical judgment, real-world experience operating at scale, and the confidence to solve problems that do not fit neatly inside an existing tool. Explore Omarchy: https://omarchy.org/ Meet the Omacom Foundation patrons: https://omarchy.org/patrons/ Read the $10 million funding announcement: https://omarchy.org/news/2026/08/omacom-foundation-funding-hits-10m/ Subscribe for candid conversations about technology, business, AI, and the work behind building things that actually hold up. Chapters 00:00 — Omarchy Quattro and the Omacom Foundation 02:23 — How Ryan joined the Omarchy project 07:11 — Selling Omarchy to a Windows and macOS user 12:15 — Speed, installation, and everyday performance 18:04 — From side project to a $10 million foundation 22:55 — Building an agentic AI operating system 26:26 — What scaling open source taught Ryan 33:29 — Release-day failures and the stable channel 40:35 — How Omarchy strengthens Oodle’s client work 45:24 — Ryan’s idea of a relaxing weekend

Websites have changed dramatically—from dial-up pages and separate mobile sites to headless CMS platforms, static site generators, personalization, and AI-managed publishing workflows. In Episode 18, Ryan and Mark look back on nearly two decades of building websites and unpack what actually matters now: choosing the right platform, designing around real user behavior, making content easy to manage, and treating accessibility and performance as core requirements instead of optional extras. They break down the tradeoffs between Wix, Webflow, WordPress, enterprise platforms, headless CMS tools, and emerging “no CMS” or AI CMS approaches. They also explain why the best technical solution can still create a bad experience—and why familiar patterns often outperform clever interfaces. The biggest takeaway: your website is the only place your brand truly owns online, and launching it is the beginning of the work—not the end. The strongest sites are measured, maintained, and improved continuously instead of left to decay until the next rebuild. If you build, manage, market, or approve websites, this episode is a practical tour through the choices that shape performance, usability, content operations, and long-term value. CHAPTERS 00:00 — Ryan’s first website and the dial-up era 02:49 — How mobile changed web design 06:14 — Mobile traffic, accessibility, and knowing your audience 10:56 — Navigation trends and the myth of the page fold 12:05 — Homepage myths, popups, and gated-content friction 15:57 — The evolution of content management systems 18:07 — Why headless CMS changes the architecture 20:36 — Performance, maintenance, and AI-friendly websites 21:24 — No CMS and AI-managed publishing workflows 23:17 — Wix, Webflow, metadata, and the human review layer 26:23 — A CMS as a central content source 27:51 — Choosing a platform by need and maturity 29:36 — Enterprise CMS, personalization, and security 31:47 — The three success criteria for a website 33:04 — Your website is the only place you own online 35:26 — Website redesign cautionary tales 36:31 — Technically correct versus usable 38:17 — Why familiar UX patterns win 41:21 — Final takeaways and future website episodes

AI was supposed to save time. So why does it feel like it is making us work more? In Episode 17 of Not Brothers, Mark and Ryan unpack the AI efficiency paradox: the strange pressure to keep agents running around the clock, maximize every subscription, and constantly wonder whether someone else has built a better workflow that is leaving you behind. Ryan talks about the practical bottleneck created when AI agents move faster than humans can review their work. Code, tasks, and deliverables can stack up in parallel—but somebody still has to verify that the output is correct, useful, and worth keeping. More production does not automatically mean more progress. They connect AI FOMO to the older workplace obsession with being busy. If the goal is meaningful impact, then free time, strategic thinking, and the ability to step away may be better signs of success than a dashboard full of running agents. The conversation also gets into token leaderboards, lines-of-code metrics, false productivity, AI-generated rework, and why companies will struggle to measure the economics of AI usage. Mark and Ryan argue that AI works best as an augmentation layer: remove repetitive work, improve the quality of the outcome, and free humans to do the thinking that actually matters. The bottom line: use AI where it produces real value. Stop using it just because it is available—and do not be afraid to shut the agents down and walk away. Chapters: 00:00 — The AI efficiency paradox 00:34 — Why faster AI can make us work longer 01:25 — AI FOMO and the pressure to run agents 24/7 02:18 — When agent output outruns human review 03:07 — Letting agents work while you step away 03:49 — AI and the new work-life integration problem 05:23 — Being busy is not the achievement 06:31 — Downtime, tinkering, and technology burnout 07:38 — Why technologists fantasize about going off-grid 09:18 — Token usage is not productivity 10:50 — Broken metrics: lines of code and AI leaderboards 13:08 — The coming economics of token efficiency 13:57 — When using AI takes longer than doing it yourself 14:46 — AI FOMO and fear of being replaced 15:46 — AI should remove busywork, not people 17:12 — A practical AI win: website content migration 18:19 — What clients actually want from agency AI use 20:25 — The cost-cutting trap and losing human expertise 21:54 — Most people are not running 15-agent companies 22:58 — Experimenting without worshipping the workflow 24:20 — Know when to stop the loop and walk away 25:11 — Wrap-up

AI loops are changing how people get real work done with agents. In Episode 16 of Not Brothers, Mark and Ryan break down what “loops” actually mean in practice — not as vague AI hype, but as a better way to structure work: define the goal, write the success criteria, pair the builder with a reviewer, and let the system iterate until the work meets the bar. Ryan shares an example of a development task that ran for hours with a builder agent and review agent working together. Instead of babysitting every step, he gave the system clear win conditions and let it loop through implementation, critique, and revision. They also talk about why this is not really new. Good AI workflows are starting to look a lot like good human workflows: clear requirements, tight scope, explicit assumptions, and a shared definition of done. The episode moves beyond software development into creative and marketing work — including social content, clips, transcripts, visual QA, and repeatable production workflows. The big idea: loops work best when they are simple, scoped, and reviewed. Chapters: 00:00 — What are AI loops? 00:27 — Why loops are not exactly new 01:10 — From prompt-by-prompt work back to real requirements 02:58 — OpenClaw, cron jobs, and early agent loops 03:35 — GPT-5.5 and goal-oriented prompting 04:55 — A four-hour builder/reviewer agent loop 06:03 — Chunking work versus letting loops run 06:44 — Good AI workflows look like good human workflows 08:34 — Why nuance and hidden requirements still matter 09:53 — The review agent is the critical piece 10:40 — Win conditions tell the loop when to stop 12:29 — The value shifts from typing to thinking 14:01 — Solving the wrong architecture before code exists 14:54 — Why knowledge work is undervalued 15:59 — The boiler repairman and the value of experience 17:13 — Value-based billing and technology leverage 19:18 — The dopamine trap of instant AI iteration 19:43 — Prototyping first, then turning reactions into requirements 20:11 — AI can help write specs, but you still have to read them 21:50 — Applying loops to creative and social content 23:34 — Getting to 90% before humans review 24:35 — Using transcripts as requirements for content loops 25:28 — Stop staring at the harness and start defining the contract 26:26 — Start with small loops, not a 50-agent circus 27:21 — Simple workflow loops like standups, notes, and PR rebases 27:53 — Wrap-up

What happens when a digital agency starts building and launching its own software products? In Episode 15 of Not Brothers, Mark and Ryan talk about the lessons they’re learning from bringing Oodle-built products like Herald and Nebula to market. After years of helping clients promote finished products, they’re now dealing with the messier front end: product, price, placement, positioning, launch decisions, feature scope, and the temptation to keep polishing forever. They dig into why products are often ready before the builders think they are, how “done is better than perfect” applies to SaaS launches, and why overbuilding can make a product harder to sell instead of easier. They also talk about the product design trap of adding too many permissions, settings, toggles, and safeguards for problems that do not exist yet — plus why simple user experiences are much harder to create as products become more capable. Later, the conversation turns to AI: how hallucinated research can quietly poison go-to-market work, why teams need better verification habits, and how skills, agents, orchestration, and daily briefs can create real productivity gains when used with discipline. Chapters: 00:00 — No topic, so let’s talk about what we’re building 00:22 — Launching Herald as Oodle’s first product test bed 01:25 — Moving from promotion into product, price, and placement 03:11 — Nebula, feature creep, and products being ready before you think 05:44 — Shipping early enough to get real user feedback 07:05 — Done is better than perfect 08:35 — Day one starts when real users touch the thing 09:48 — Finding and removing features nobody actually uses 12:21 — Don’t add complication until it is necessary 13:03 — Permissions, theoretical problems, and soft safeguards 16:59 — Settings, toggles, and exposing complexity in the right place 18:24 — Why simple UX keeps getting harder 21:24 — Eating our own dog food while building products 22:18 — Using AI for market research without accepting bad data 24:10 — AI is not a Google search result 26:34 — Skills, repeatable workflows, and progressive disclosure 29:46 — Running multiple AI sessions without losing the plot 31:45 — Orchestrators, review agents, and long-running AI work 34:10 — Applying agent workflows beyond development 35:32 — Daily briefs, AI loops, and reclaiming focus

In this episode of Not Brothers, Ryan and Mark dig into a major shift inside modern teams: AI now lets non-technical people prototype, build, and express product ideas in ways that used to require developers, designers, and long handoff cycles. That's powerful. It's also messy. The upside is real. AI can take someone from a vague idea to an interactive prototype incredibly quickly, compressing wireframing, design, and prototyping into a tighter feedback loop. Designers, strategists, and PMs can create something tangible enough for the team to test and improve. But a slick UI can create the illusion that something is "done" when there's no real infrastructure, no secure backend, and no maintainable architecture. AI is great at making something that feels real — but often it's a house of cards no responsible team can simply deploy. The team shares what Oodle has worked through internally: oversized pull requests, skipped requirements, one-off solutions, spaghetti code, missing docs, and unclear handoffs. The big theme — AI doesn't remove the need for product thinking, it makes it more important. Ryan's example: flexible custom fields in Cortex beat hard-coding for one client. And his best metaphor: AI will bore through a concrete wall with a spoon if you ask it to, so planning still matters. Rather than banning the tools — unrealistic, since "life finds a way" — Oodle builds guardrails: project instructions, agent rules, standards, and an internal assistant, Sheldon, that asks clarifying questions and turns vague bugs into actionable reports. Handoff quality matters too. If you build something with AI, you still own it: explain the problem, document intent, set success criteria, and make review easy. The workflow is also inverting — technical people now ask non-technical teammates to carry prototypes further before development takes over. The takeaway is simple: write things down. Clear writing, intent, and documentation are becoming core skills for turning ideas into real software. Chapters 00:00 — The collision between technical and non-technical teams  01:11 — AI gives non-technical people a new way to express ideas  03:12 — Why “done” is harder to define now  05:34 — When a polished UI creates the illusion of progress  07:46 — Why AI prototypes often are not deployable  11:20 — Guardrails, standards, and responsible AI workflows  15:02 — Existing products make the collision more complicated  17:22 — Product design versus one-off feature requests  20:25 — AI will dig through concrete with a spoon  22:05 — The problem with huge AI-generated pull requests  24:47 — Why smaller chunks beat massive code drops  27:57 — Compression, cleanup, and maintainability  30:28 — Better pull requests need demos, screenshots, and context  36:01 — What open source is teaching us about AI-generated code  40:02 — Why banning AI is not the answer  42:31 — How agents can improve bug reports and feedback loops  45:17 — Don’t clean up everyone else’s AI mess  49:09 — The workflow has flipped for idea people  52:59 — Writing clearly is now a core AI-era skill  55:10 — Final takeaway: write it down

Short podcast summary Mark and Ryan dig into Herald, Oodle’s developer-native changelog and release notes platform built for teams that ship through GitHub but hate writing product updates from scratch. Ryan explains the gap he found in existing changelog tools, why release notes usually get skipped, and how Herald uses GitHub history plus AI to turn commits and pull requests into editable release drafts. They also cover GitHub sync, nested projects, scheduled releases, customizable widgets, email notifications, and user segmentation — all aimed at making product updates easier to publish and easier for users to discover. YouTube description Most teams ship more than they communicate. In this episode of Not Brothers, Mark and Ryan talk through Herald — Oodle’s changelog and release notes platform for software teams that live in GitHub but hate writing release notes from scratch. Ryan explains why changelogs are usually skipped, why existing tools did not quite fit the workflow he wanted, and how Herald turns GitHub activity into draft release notes using AI. Instead of starting with a blank page, teams can connect a repository, pull in commits and pull requests, draft a release, edit the important parts, and publish across Herald, GitHub, email, and an in-app widget. They also get into two-way GitHub sync, public and private repositories, nested projects for related repos, scheduled releases, customizable changelog widgets, user groups, segmentation, and why discoverability matters just as much as authorship. Herald is built for developers, product teams, indie founders, and small SaaS teams that want to keep users informed without turning release notes into another full-time job. Try Herald: https://sendherald.com Chapters 00:00 — Why Oodle built Herald 00:44 — What Herald is and the changelog problem it solves 03:02 — Release notes for users, engineers, and bigger feature launches 04:56 — Using AI to turn GitHub activity into draft changelogs 06:21 — Moving from creator to editor of release notes 07:22 — Two-way GitHub sync and avoiding duplicate work 09:31 — Custom categories and tuning the AI import prompt 10:25 — Public/private repos and nested projects 11:31 — Multi-repo product families and parent changelogs 13:08 — Scheduled releases 14:22 — Getting started without a blank canvas 15:30 — Drafting a release from everything since the last GitHub release 16:43 — Customizable in-app changelog widgets 17:36 — Making product updates discoverable 19:28 — In-app updates vs. noisy notifications 19:59 — Groups, JWT, and segmented changelog visibility 21:44 — Internal users, client users, and beta release use cases 22:10 — A simple tool that adds value in the right capacity 23:07 — The three user types Herald is built for 23:48 — Real release notes, testing, and future feedback 24:35 — Website demo and interactive examples 24:59 — Try Herald and let us know what you think

Description The AI gold rush is hitting its first real hangover. In Episode 12 of Not Brothers, Mark and Ryan talk through the gap between what AI companies promised, what executives bought into, and what the tools are actually proving they can do. The conversation starts with cloud-license cancellations, token spend, AI data-center bets, and the realization that “AI will solve everything” is not the same thing as a useful operating plan. Ryan argues that AI is still an incredible tool — even if it never gets dramatically smarter — but the fantasy of universal automation, effortless AGI, and instant economic transformation is starting to crack. Mark pushes on the business side: why executives accepted the hype, how fiscal pressure may be changing the story, and why the next phase of AI value may come from practical application layers instead of frontier-model moonshots. They also get into AI dopamine loops, hallucinated research, agentic coding tools, the iPhone analogy for model progress, Sam Altman softening job-replacement claims, data-center and memory-market ripple effects, Google’s AI distribution advantage, Google Workspace integration, and what AI search might do to SEO. The takeaway: AI is not going away. The useful version is probably less magical, more embedded, more specialized, and much more dependent on human judgment than the hype cycle promised. Chapters 00:00 — The AI economics hangover 01:24 — Executives, overpromising, and shareholder-value promises 02:40 — Why AI hype is easy to sell upstairs 04:30 — Token drunkenness and the cost reality check 05:54 — Fiscal pressure, Microsoft, Claude, and Copilot 07:26 — Finding the limits of agentic AI tools 09:44 — Goalposts, model progress, and AI fatigue 11:55 — The iPhone analogy for frontier-model improvement 14:18 — AGI goalpost shifting and useful-but-not-magical agents 16:49 — Model economics and better autonomous coding loops 18:26 — Dopamine machines, fake confidence, and verification 20:48 — Reddit, authenticity, and trust in AI training data 21:56 — Sam Altman, job disruption, and the softer economic view 23:29 — Is AI a bubble or an early overbuild? 24:38 — Data centers, memory prices, and supply-chain ripples 26:48 — Infrastructure bets and consumer/app-layer demand 29:03 — Google’s distribution advantage in AI 30:02 — Gemini, coding models, and different model strengths 31:04 — Google Workspace as the AI surface area 32:34 — AI search, generated answers, and SEO disruption 33:20 — Actual content people want may finally matter 35:36 — The echo chamber vs. mainstream adoption 36:33 — Untapped users and the application layer 37:39 — AI inside existing tools, not only standalone chatbots 38:02 — Better chatbots would still be a win 38:32 — Wrap-up Pinned comment / hook AI is still powerful. The fantasy version is what’s getting repriced. Tags/topics AI, AI economics, AGI, token costs, AI agents, OpenClaw, OpenAI, Anthropic, Google Gemini, Google Workspace, AI search, SEO, data centers, jobs, automation, future of work, Not Brothers Podcast

AI can make building products faster. It does not make people care. Distribution, trust, and attention are still the real game. Description Building software is easier than ever. Getting anyone to care is still the hard part. In Episode 11 of Not Brothers, Mark and Ryan dig into the modern version of “if you build it, they will come” — and why that idea breaks down fast in an AI-driven product world. Vibe coding, faster prototyping, and smaller teams have made niche software products more realistic than they used to be. But the same tools also make it easier for competitors, clones, and half-baked alternatives to show up overnight. The real debate: has the power shifted from developers to distributors, or was distribution always the thing that separated products that survived from products that disappeared? Mark and Ryan talk through AI-era distribution tactics including AI-friendly tools and CLIs, MCP servers, programmatic SEO, answer-engine optimization, free tools, shareable product outputs, niche newsletters, cold email ethics, and content repurposing engines. Along the way, Ryan gets predictably fired up about MCP bloat, AI slop, automated outreach, and bots talking to bots until everyone involved is just burning tokens. The takeaway: AI can help you build faster, but it does not magically create trust, attention, demand, or distribution. If you build it, they probably will not come — unless you give them a damn good reason to. Chapters 00:23 — Why AI changes the product-development conversation 01:08 — Has the Silicon Valley pecking order flipped? 02:27 — Distribution was always the hard part 04:20 — When product moats get easier to copy 05:04 — Salesforce, Oracle, and the power of incumbency 06:41 — Niche products in the AI era 07:16 — Why small markets used to be hard to serve 09:38 — The new case for niche software businesses 10:23 — Using AI-friendly tools for distribution 11:05 — Ryan's problem with MCP servers 12:42 — Distribution paths beyond MCP 12:54 — Programmatic SEO and the slop problem 14:02 — LinkedIn, AI content, and the loudest voice in the room 15:58 — Answer-engine optimization vs content spam 17:10 — Good old-fashioned inbound marketing, now AI-readable 19:10 — Free tools as top-of-funnel distribution 20:17 — Why interactive tools build brand equity 21:10 — Making product outputs shareable 22:28 — Buying niche newsletters and owned audiences 23:27 — Ryan draws the line on spam 25:27 — AI agents, cold outreach, and inbox overload 27:27 — When sender bots meet screener bots 28:17 — Why AI does not belong in every communication layer 29:56 — AI content repurposing engines 32:07 — Using AI to extract the useful five minutes 33:44 — Social volume, quality, and the For You page 35:27 — The final answer: building is easier, distribution still wins

Does AI make us dumber, or does it just move the line between what humans need to know and what tools can handle? In Episode 10 of Not Brothers, Mark and Ryan pick up the thread from their previous conversation about AI in education and push it further: if AI can write the paper, build the CLI app, summarize the research, and automate the busy work, what exactly are humans supposed to learn, practice, and protect? The conversation gets into education, critical thinking, memorization, work, hobbies, purpose, the future of AI adoption, and the difference between delegating execution and outsourcing your brain. The short answer: yes, AI can make you dumber at the thing you delegate. But that may be fine if you’re using the saved time and leverage to get smarter about the thing that actually matters. 00:00 — Does AI make us dumber? 01:02 — AI in education vs AI at work 02:27 — Delegating your brain 03:44 — What are schools actually measuring? 06:04 — Real-world skills vs academic restrictions 06:54 — Resourcefulness vs intelligence 09:23 — AI, memorization, and what we call “smart” 10:31 — Does learning need to be hard? 12:26 — Are we at another inflection point? 14:23 — Human purpose when work changes 16:23 — Universal high income and building for fun 18:00 — Retiring without a backup plan 20:25 — What do we do with AI-created time? 21:17 — Are AI models plateauing? 24:08 — The IKEA example: AI plus human judgment 25:43 — Acceptable AI use in education 27:48 — When does AI work become “mine”? 29:42 — Building blocks and the 10-year-old problem 31:07 — Handwriting, typing, and obsolete skills 33:51 — Brain development and hard things 35:52 — Why AI adoption feels faster than the internet 37:53 — Can AI or the internet be regulated? 40:15 — So, does AI make us dumber? 40:30 — Dumber at one thing, smarter at another 42:45 — Critical thinking vs subject matter expertise 44:18 — AI is best at patterned execution 46:27 — The final answer: maybe 48:24 — Are papers even the right test? 50:06 — Education needs to figure this out