gen alpha + captain fantastic

Z / Runner 🏃🏻‍♂️

🏃🏻‍♂️ tgi-"gen-alpha"; past 50 years, safe path: STEM major, ivy league, SWE jobs (stem no longer moat, learn to code = table stakes at best, negative roi, lower signaling premium, no career is safe, swes, lawyers, doctors).

• Parents = we all feel it, just don't want to admit it (worked well for gen x & millennials)

• Kids = initiative, agency, judgment, the ability to create value from nothing.

• Core = not about "better parenting", the soft stuff. We focus on moving kids from passive curriculum to active production. Parents open the door, kids must walk-thru.

• History = Medicis, Rothschilds, Morgans, Rockfellers, Waltons, Trump, Musk… all expose kids to power & apprenticeship. Why not replicate for the middle class? A new type of upward mobility in the era of economic singularity.

• Ops = age 7+, parents/kids co-build one project per month. Real projects. Real pitches. Real stakes. NOT science fair posters, fake lemonade stands. End of each month: a mini YC pitch day. Kids presents, parents question. Capital allocated. $200 fees per month, prize pool for 1st place (n1) = $1000; 2nd place (n2) = $100; 3rd place (n3) = $10.

• 10x better than after school tutors?

• 🏃🏻‍♂️ public speaking, presenting = learning to influence, communicate, tell stories, important; pitching = learning to fail and recover (eileen gu, evidence over affirmation)

• 🏃🏻‍♂️ justify, i strongly believe in rewards/incentives…

• 🏃🏻‍♂️ i want real business for kids, not cute. 🏃🏻‍♂️ exactly, but NOT a fair… consistency > intensity; practice > game, business > fair.

🏃🏻‍♂️ Q1 is done, killed longevity cafe in jan (too ops heavy, logistics drag). This week, go/no-go on tgi-"xyz". 96 days of daily peer-session, is this replicable for 10 groups of "n"?

• Context: local biz got me thinking. What about a monthly parents + kids builder meetup? We used to have bring kids to work at google (no longer the case, security reasons on campus). Think of it as a simple projects, weekend-friendly, 7 mins demo/pitch. Lightweight, low overhead (borrowed space). Worth testing? Yo & I are regulars at this place (taste buds). Worth testing?

• 🏃🏻‍♂️ YES. charge $42 per session, food & drinks separate. This is an "anit-social" social club. Kids need to work, setup, clean up, pre-meeting… demo.

• 🏃🏻‍♂️ getting extremely mixed patterns on the topics of shopping. All the tech "bros" conversation = "i don't shop"... talking to other parents at school & family members, they all wish AI can help w/ shopping. Cost saving being the top reason & recommendation being the close second.

• 🏃🏻‍♂️ I burn thru 1 pair of running shoes every 4 weeks ($200 - $220 per pair); most marathons are $400+ (excluding hotels)... looking at my credit card bills (the NVM [🏀 nevermind? no acronym!] ~= $1700)... it adds up very quickly. 🏃🏻‍♂️ napa valley marathon (NVM). 🏀 !!

• 🏃🏻‍♂️ I want to be very careful & nuanced here: i am NOT saying what we are building is a shopping agent, I'm saying the immediate revenue potential is the obvious/obvious… and so far I have NOT heard any logical counter-arguments. This is the $100K MRR you were asking last year.

• 🏃🏻‍♂️ yeah, fallacy on "n of 1"; i don't eat meat, doesn't mean most don't :-)

• the 10x gain is in discovery – search (crawl + index + rank) is fully broken across all verticals: youtube, substack, twitter. i can take recommendations only via friends on their trusted domains – along with wirecutter, american test kitchen. 🏃🏻‍♂️ fan of americastestkitchen.

• 🏃🏻‍♂️ 100% agree… hence, I think extraction + synthesis = level 3. I understand ai summarization sucks, but i do think most people don't the time & the motivation to ingest YT, SS, X, Reddit… our inner circles are unique b/c we have time (for the time being). But our style of content consumption isn't the norm.

• 🏃🏻‍♂️ old man dying… content creators / taste makers always talk about how "they are special"... people need to get off their pedestal. AI content generation will kill 90% of "Rick Beato". Example here, will this win the next emmy, no; will millions of parents do this for fun, yes. This is where creatives don't get tech, and tech people don't understand creatives. 2+ years ago, many software engineers are so sure of their future…

• 🏃🏻‍♂️ that's NOT the point of these, it is "plastic forks" of content. Use it once and get a new one next time.

• 🏃🏻‍♂️ 2nd order effect, we should do food b/c it "decorates the vibe", if we make this into the core, we are DOA. For example, if we do matcha b/c it is part of the longevity cafe, then ofc.

• 🏃🏻‍♂️ → experience economy = needs to be kids friendly and 0% regrettable feelings after

• 🏃🏻‍♂️ took me 3+ weeks to decide which super trainer shoes to buy… all the product launches, reviews, recommendations, feedback, discounts are embedded inside of youtube these days.

• 🏃🏻‍♂️ 100% agree.

• 🏃🏻‍♂️ we need to launch an app… hence, been late night vibe prototyping.

• 🏃🏻‍♂️ agree… healthier mindset, suck 1% less everyday, hence consistency > intensity. How many times have we heard or done something that fits this "虎頭蛇尾"... 99% of vibe coding is just performative public ego striking …

• 🏃🏻‍♂️ millions of blogs in the early 2000s, only a few are willing to do the work (ben thompson, john gruber…) mid 2010s, millions of youtubers and only a few are willing to do the work (MKBHD), millions of podcasters and only a few are willing to do the work, millions of vibe coding performing SWEs… few are willing to do the work.

• 🏃🏻‍♂️ ICP [ideal customer profile] 1 (me): early 40s, 1st time parent, tech/product background. Builder first, operator second; bi-culture/lingual (west ←merge→ east). Past life = local optimization, career to life-long mission. Now = global max, system-level tradeoffs are visible, rejects incremental gains, rejects past patterns, rejects popular sentiments. Challenge = coherence across health, wealth, family, work; no longer believes in "great businessman, but asshole".

• 🏃🏻‍♂️ yes, respects old guard, questions inherited lessons, only want to update priors if 1st principled. Still in love with tech, but abhor hypes. Pain = focus frag, time compression post-child. Silent drift of "say-do" gap, falling off the bandwagon is the BIGGEST weakness.

• 🏃🏻‍♂️ ICP 2 = age 39, 1st time mom, child under 5. Born + raised in asia, 20+ years in the US. More than ½ of life in US, native-born of globalization (Eastern culture fading, western culture growing). Well-educated, graduate degree common, career in tech-adjacent role, pms ops, strategy. Tech as profession. Not childhood identity. Ambitious, but practical. Internally driven, not status seeking. High agency, low noise tolerance. tension = retain eastern values, but operates in western system. Willing to slow career for child. Conscious sacrifice, not default. No resentment, but quiet recalibration. Used to optimize for achievements for school, work, music, but optimize for presence. Love = still loves high culture, books, classical music, jazz, michelin dining, design, fashion, architecture; now framed as indulgence, time scarcity creates guilt. Curiosity compressed. Not fitness lazy but reality constrained. Consumes depth, not news. Reads mandarin + english sources. Understand both narratives, trust data. Rejects shallow empowerment content, hustle cliches. Pain = loss of intellectual space, identity frags, fear of silent self-erasure. Will NOT pay for productivity tool, influencer noise, surface-level features.

Stephen / Basketball 🏀

• 🏀 captain fantastic. "un-schooling or un-parenting, ideally in a forest like walkabouts, is what i hope to gift my kids. a non-greedy algorithm for exploring the future – "don't peak too early". so much struggles and opportunities, don't optimize for the rules." 🏃🏻‍♂️ <3

• 🏀 not pitch or hackathon or prize – directly to creating value and collecting revenue – a pico-biz via ai agents, shared scaffolding, everyone else's help.

• 🏀 mostly digital services, selling to parents + kids in attendance or in school. e.g. organizing (math) competitions, portrait drawing, handcraft crochet, selling matcha ice cream. per kidizens (emma and etty's fav class).

• 🏀 i will bring 4! but that's strictly social / for kids. for demos, either we need sponsors or charge $100.

• 🏀 z, what time / cost / impact % can your grocery shopping be of manual vs agent? shoe / children / fashion.. shoppings %?

• 🏀 saving 10% for shoes, 5% for grocery.. race to 0?

• 🏀 except i don't use any / shop. probably just wanted 2x more variety on grocery lists.

• 🏀 100x opposite take; per Rick Beato: AI generated..'Ugh, boring, boring, boring.' And my kids do the same thing.

• 🏀 hm, audio / dance music ok-pass for a 5-year-old; not tomorrowland level.

• 🏀 beer + ramen → matcha + _.

• the 10x cheat is in time – no need to rush to give an definitive answer; filter 1m products / places, 1000 reviews on each of the top 10, compare to contrast the tradeoffs.. before answering in a few hours. currently gemini cites low-quality reddit posts, 100-view-count or 5-year-old videos youtube .. even recommending permanently-closed shops!

• 🏀 the opposite for me, as you noticed. probably only at the same (wrong) level of obsession on food, films, _.

• the 10x fail is in corpus – the top 10(00) results are spam; must go fully exhaustive. try "10 best matcha desserts in berkeley ca" on your ai prompt / multi-round sessions / manus agents to find my ground-verified fav yume mori matcha, chicha san chen 吃茶三千; ditto "restaurants" for noodle dynasty.

• the 10x ux is in progressive – always work further on the task for 12 hours max; use unlimited retries. return via fast mode in 1 second, pro mode in 30 seconds, deep think in 5 mins, agent mode in 15, 30, 60, 120, 240, 480.. doubling mins – all in the same browser tab with auto refresh.

• 🏀 but are we so afraid of making apps / sites / wearables..?

• 🏀 sadly, 1000x more people is vibcoding 90% ok product, not 99.9% great code anymore.

• 🏀 but that's 100% on consumer digital (ai); the rest on enterprise are just in-loop serving to those (pointless) consumer digital needs?

🏀 "top 3 highlight segments of this podcast for a 48-year-old compiler phd, map search + crypto web3 + agent startup founder, cantonese american in palo alto ca, 4 daughters under 13; daily basketball + films, weekly beach swims, monthly michelin stars; worked at google / apple / microsoft. cite the full original passages and add a link with ?t=timestamp"

• z, what other bio / background / factors in making the cuts?

• 🏀 also: daily healthy vegetarian + runners, monthly marathons.. which other attributes you'd add for me as prompt / context to have a good podcast highlights?

• 🏀 ah, i lost (most of) my love for math / tech. ha, we probably need more of a coach than mentor / advisor / agent / recommender.

• 🏀 yo's profile? what intellectual space?

• 🏀 "what other questions or preference reveals on a person in order to personalize the best highlight clips from podcasts?" – listen to an hour at 1.5x speed during self-driving and zone 2 shoot around. daily 90 mins reading of hacker news, wsj, nytimes.