The Good Stuff, with Pete and Andy - Episode 39: The Big Stuff

Hosts: Pete and Andy

Pete accidentally calls it "The Big Stuff" and decides to run with it. They dive into Dumpling Town—a game Pete built with his daughter in 90 minutes that now has an economy, side quests, and soy sauce rivers. Treating agents like humans, building family chat apps to avoid WhatsApp, the new Hal orchestration agent, and reflections on a year of shipping software.

Key Moments:

[01:48] Building Dumpling Town: Animal Crossing meets dumplings.

[03:43] The creative genius of kids: "Why don't kids make all of the games?"

[04:48] Level Up - Touch Don't Look for schools.

[08:44] Starting with Nostr: get logins and authentication for free, no password recovery needed

[09:53] Built a Nostr-based gratitude journal: encrypted, runs in the house, wife can use it too

[11:40] The Nostr database assumption: you don't have to store everything on relays

[12:51] Full encryption in the Slack competitor—can't see database data without the key

[15:05] Treating agents like humans: they're just npubs in the room, add them to groups or don't

[16:10] Andy's approach: "I treat them very much like I treat my humans—mush, mush, do it now"

[16:44] Health graph experiment: building while talking to Claude, using sub-agent for graph construction

[18:32] Interface matters: weird to have deep conversations in the terminal

[19:57] Marginal gains as the pre-planning space for Wingman—natural workflow emerged

[24:07] Health-specific interface: makes sense to have constraints around what agents do

[25:03] Dumpling Town expansion: everyone gets their own town, planes between towns, Olympic Dumpling Island

[26:42] This is invaluable education for young kids—unlimited creativity meets approachable tools

[28:00] Schools could run their own Wingman—teachers should love it

[29:15] Learning from first principles: hand-coding microprocessors in binary teaches abstraction value

[31:00] Should people still learn to code? "What do you mean should? People who want to will"

[32:27] Be the Japanese woodworker, not the IKEA table producer competing with robots

[33:07] Gigi's realization: "Fuck, it's fun again—now I can create, I just think things into the world"

[38:06] Built family chat app: encrypted, data lives in the house, kids can message without WhatsApp

[41:21] The one-shot fallacy: Twitter theatre vs actual building over two weeks

[44:44] The container for AI is the business, not the app—software serves the business

[47:00] Re-win amplification: personalized software that does exactly what you want

[48:52] Family as a business: shared calendar, action lists, silly memes channel, pocket money with wallets

[51:26] Mission accomplished for 2025: developed the muscle for shipping software

[52:32] GitHub tracker shows the progress: scattered commits to daily by end of year

[54:29] The wrong framing: doesn't matter if you can't one-shot—two weeks is still 1000x faster

[56:52] Pay-per-day pricing model: 21 sats a day, buy a week and see how it goes

[59:44] Starter kit idea: preload authentication and payment interface for rapid deployment

[1:00:02] Introducing Hal: named after Hal Finney, orchestrates and manages all running apps

[1:03:12] Level Up: the game workshop name—sounds right, move on

[1:05:07] MCP with Blender: could solve the Dumpling Town sprite problem

[1:07:09] Look Marks: Gigi's app that treats eye emojis as bookmarks—because Nostr is open

[1:08:00] Christmas phone detox: collapsed screen time, more quiet deliberate time

[1:09:41] Raw dogging everything: no music, no podcasts, just silence—harder than expected

# The Good Stuff, with Pete and Andy - Episode 38: The Rise of Vibe Product Manager

**Hosts:** Pete and Andy (New Year's Eve edition at the beach)

Pete reveals he's stopped coding entirely after 12 months of learning—now he just product manages a team of agents. They explore the shift from "vibe coding" to "vibe product management," why Pete built Slack in a day, and the addictive power of tight feedback loops.

Plus: steel-manning the security concerns, why raw dogging agents beats sub-agents, and New Year reflections on why model breakthroughs don't matter anymore.

## Key Moments:

* [02:07] "I learned to code for 12 months. And then I just stopped. It's nuts."

* [02:54] Pete built Slack/Discord in a day: "I shouldn't be able to replace a $20 billion SaaS company in 12 hours"

* [08:10] The addiction: "From nothing to MVP in a day, radically transform it every morning"

* [09:05] Not vibe coding—it's a glorious tool for entrepreneurship and building businesses

* [12:10] Introducing Agora: Slack replacement with integrated Kanban, all Nostr-based and encrypted

* [13:25] Slack's fatal flaw: "Bullshit asynchronous tool because you just lose everything"

* [18:05] Steel-manning the critics: addressing "you'll get hacked" security concerns

* [21:21] Everything encrypted with Nostr keys: "They literally can't steal anything"

* [23:21] Your attack surface: "You're the oil you get out of a peanut—not worth it to hackers"

* [28:03] Raw dogging the base agent—not bothering with sub-agents or elaborate skills

* [31:44] Pete's tried sub-agents many times: "I have always been left wanting"

* [32:33] The breakthrough pattern: specify in 10 minutes, deliver in 5-10 minutes

* [34:15] The 21-minute rule: must complete the loop in 21 minutes or it doesn't work

* [42:04] Mainstream adoption reality: experienced professionals never heard of coding agents

* [47:29] Small business owners are the right audience—different relationship to risk

* [50:09] Small businesses always full of risk: "Your hair is always on fire anyway"

* [51:56] The cottage software developer—1000 true fans model, not billion-user SaaS

* [56:44] Andy's New Year take: bullish on tools, no model breakthrough needed since 3.5

* [57:33] Touch Don't Look success: normal people cross the hurdle in 10 minutes, creativity shines through

**Quote:** "I learned to code for basically like 12 months. And then I just stopped. It's nuts. And I thought, this really is like the way to use these things."

# The Good Stuff, with Pete and Andy - Episode 37: Stocking Fillers - Predictions for 2026

Hosts: Pete and Andy (sporting festive headgear at the beach)

Pete and Andy kick off their Christmas predictions episode with zero preparation and maximum confidence. They tackle the biggest question: what will happen in 2026?

Key Moments:

[00:51] Andy's two-for-one Christmas prediction: most AI predictions won't eventuate, no AGI in 2026

[03:09] Revised timeline: feels more like a 10-year transition than 5 years

[04:02] Pete's first prediction: models become less relevant, focus shifts to agent tooling

[05:40] Are model releases becoming underwhelming? The breakthrough isn't as significant anymore

[05:53] Elon's take: "The whole framework is incorrect—throwing more data doesn't give you intelligence"

[07:08] Why software automation is so valuable: automate software, you get free option on everything else

[08:17] Less discussion about new model releases—when Opus came out, people raved for a few days then moved on

[09:05] Death of one-shot benchmarking: "Nobody uses these things in that fashion"

[10:26] The Presidio Bitcoin example: Gemini 3 got everything wrong because the task was too broad

[12:07] Flow matters: sequential tasks build on previous work, jumping around creates cognitive load

[13:15] Andy's prediction: website development will be completely commoditized

[14:01] Big agencies won't be able to justify high spends anymore—except for enterprise clients

[16:24] The WYSIWYG problem: hard to update flat file websites without developer tools

[18:13] Pete's prediction: agents move out of the terminal

[19:08] The model switching advantage: when one gets nerfed, quickly switch to another

[20:30] Why Goose is better for non-coding tasks: less sandboxed, happier to just do work

[22:06] The spare machine problem: agents need machine access to be powerful

[23:47] Most people still talk about Copilot, not Claude Code or Codex

[24:44] The Excel analogy: "This is like super Excel, why wouldn't you want to learn it?"

[27:06] Calling them "coding agents" creates mental resistance for non-coders

[27:27] The agent becomes the engineer, you become the product manager/technical architect

[30:46] The killer use case isn't software—it's the business you already have

[31:20] Service 10x the market with no additional headcount: "That's insane"

[32:29] Value will accrue to small businesses, not S&P 500 companies with cultural inertia

[33:50] Andy's prediction: AI becomes more politically negative in 2026

[34:36] Politicians will use fear-mongering based on job loss to accrue more power

[36:43] Greater divide between adopters and laggards—companies that embraced AI take big leap forward

[38:38] Small nimbler SMEs will be the standout stories, not big enterprises

[39:32] More of your business can run in software than you thought it could

[40:40] The Replit debate: "Why would I use Replit when I can just change Wingman myself?"

[41:40] Wingman needs the concept of a business inside it—metadata that flows through apps

[44:26] Pete's hot take: large models are a dead end, we don't need bigger models with more data

[45:20] "The model should understand what you're doing, not be responsible for knowing stuff"

[46:40] Fast models over big models: "I think that becomes the rallying cry"

[48:02] The inverse direction: everyone's been focusing on thinking time, slower and more deliberate

[49:12] Speed unlock: "If it took 10 seconds instead of 10 minutes, you'd use it 100 times more"

[51:15] Pete's fundamental belief: "I don't think you want the model to know stuff—it's a bug we ship as a feature"

[52:56] Why domain-specific models don't make sense: graphs do the heavy lifting of knowing

[55:05] Timeline check: specialized models and speed focus probably not a 2026 thing

[57:24] Rise of the Agents: more use, simpler to use, non-coding use cases becoming clear

[58:38] Final agreement: agents are the future, not bigger models