Max Schoening (Head of Product, Notion)03 May 2026

Why cultivating agency matters more than cultivating skills in the AI era

7Frameworks

15Insights

Episode overview

Max Schoening argues that as AI makes skills broadly accessible, agency—the belief that the world is malleable and the willingness to change it—becomes the greater differentiator. He explains why designers and product managers should prototype in code to understand the medium, while preserving specialist engineering and design craft. The conversation also examines malleable software, AI's effect on product development, cultivating taste through repeated feedback, and the tiny exceptional core shared by enduring products.

Key ideas

AI makes the early portion of projects dramatically cheaper, enabling teams to replace speculative memos with working demos and explore more paths.
Agency matters more as skill barriers fall; people build it by making, tinkering, and learning that the systems around them can be changed.
Designers and product managers benefit from coding primarily because it teaches them the material and agent loops they are designing, not merely because it increases production output.
Malleable software should serve the interests of its users and give them ownership without forcing everyone to maintain an entire software stack.
Role convergence should not erase specialists responsible for reliability, scale, design craft, and the hard final portion of production work.
Taste develops through many iterations with feedback and exposure to excellent work, allowing someone to predict how a chosen audience will respond.
Great products tend to have one exceptionally strong core capability; adding more features cannot rescue a weak core.
AI's strongest near-term organizational effect may be spreading software creation into every function rather than independently transforming each profession.

Transcript available · source text is retained privately and is not published

Frameworks in this episode

Marketing3 steps

Be The User Reset (Jobs-to-be-Done)

Zoom out and ask what the user hires your product for — then be that user and ask if you'd even buy what you made.

Read →

Self-Mastery4 steps

Cultivate Agency, Not Skills

When AI hands everyone the skills, agency becomes the only differentiator — and you build it by making things.

Read →

Productivity3 steps

Demos Not Memos

The first 10% of every project is now free — so build something to react to instead of writing documents.

Read →

Innovation3 steps

Design in the Material

PMs and designers should code — not to ship, but to master the material and truly understand what they're designing.

Read →

Strategy3 steps

Obviously Good, Then Incremental Correctness

Only make obviously good stuff, ship it in iterations, then reconcile the sprawl back to a naked robotic core.

Read →

Self-Mastery3 steps

Taste as a Trainable Model

Taste is a virtual machine in your head that predicts whether your in-group will like an idea — built by reps with feedback.

Read →

Innovation3 steps

The Tiny Core Principle

Every enduring product has one tiny thing that is a superpower — find it, protect it, and stop bolting on features.

Read →

People & resources mentioned

Attributed to the moment in the episode. Timestamps are approximate.

People · 27

Nat FriedmanMentionsmentioned
“He was a design leader and an engineer at GitHub under Natt Freriedman.”
▶ Watch at ~1:00approx.
Max SchoeningMentionsHe was a product manager at Google. He ran the design team at Heroku. He was a design leader and an engineer at GitHub
“He is one of the most successful AI forward product leaders out there”
▶ Watch at ~1:30approx.
Lenny RachitskyMentionstoday we've got another very special compilation episode something I've been pulling on more and more with the podcast and the newsletter
“Max, welcome to the podcast.”
▶ Watch at ~1:30approx.
Bret VictorCoinedthis guy named Brett Victor who's kind of like a prot prot typer thinker
“there's this great talk by Brett Victor uh stop drawing dead fish”
▶ Watch at ~3:30approx.
Brian LovinRecommendsmentioned
“someone like Brian Leven, who you should probably have on the podcast at some point.”
▶ Watch at ~12:00approx.
Eric LuMentionsmentioned
“Eric Lou is another one.”
▶ Watch at ~13:00approx.
Steve JobsCoinedSteve Jobs is just uh the bar he held for the company and for technical talent and for excellence was not uh wavering.
“the thing that I care most about is the Steve Jobs quote.”
▶ Watch at ~16:00approx.
Jeffrey LittMentionsmentioned
“I get to work with Jeffrey Lit every single day now”
▶ Watch at ~20:00approx.
Brian CheskyMentionsBrian chesky at Airbnb there was a whole article where we unpacked this moment in their product development cycle
“It's something that I learned also from Brian Chesky at Airbnb.”
▶ Watch at ~20:30approx.
Dieter RamsMentionsmentioned
“which I think is DA Roms. Is that who that who the person is?”
▶ Watch at ~21:00approx.
Frank GehryMentionsmentioned
“you would think it was Frank Giri and like all these famous designers pieces put up in a museum”
▶ Watch at ~22:30approx.
Stewart BrandCoinedmentioned
“Stuart Bran's uh sort of how buildings learn I think idea”
▶ Watch at ~23:00approx.
Bret TaylorCoinedHe co-created Google Maps, co-founded FriendFeed and Quip, held senior roles at Facebook and Salesforce, and co-founded Sierra.
“Brett Taylor says this too. Software is like a garden.”
▶ Watch at ~25:30approx.
Joanna SternMentionsa journalist
“Joanna Stern um a journalist recently tweeted something along the lines of oh thanks to notionai I finally understand and use notion.”
▶ Watch at ~26:00approx.
Cat WuMentionsToday my guest is Kat Wu, head of product for Cloud Coding who work at Anthropic.
“I just had the head of product for cloud code on the podcast Catwoo”
▶ Watch at ~27:00approx.
Tyler CowenMentionshe has this awesome blog marginal Revolution and he's really good at sharing insights from research papers
“I think Tyler Cowan says something similar. I don't want to put words in his mouth.”
▶ Watch at ~34:00approx.
Simon LastMentionsIvan Zhao’s Notion co-founder and rebuilding partner.
“something that uh Simon last talks about a lot”
▶ Watch at ~38:00approx.
Kathy SierraMentionsmentioned
“I always think about Kathy Sierra.”
▶ Watch at ~47:30approx.
Ivan ZhaoMentionswhen notion needs to broadcast messages in real time to other collaborators they use a real-time engine that they built for socket io.
“I think both Ivan and uh Simon had the intuition of hey this is going to change a lot of things”
▶ Watch at ~49:30approx.
Dario AmodeiMentionsI was sitting next to Dario yesterday and he's like I keep making these predictions and people keep laughing at me
“Dario is that he wasn't Oh, he wasn't just lucky once at OpenAI.”
▶ Watch at ~1:04:30approx.
Dario AmodeiMentionsI was sitting next to Dario yesterday and he's like I keep making these predictions and people keep laughing at me
“the thing that I find the most impressive about uh I don't know who to give credit, but like obviously you give the CEO”
▶ Watch at ~1:04:30approx.
Clayton ChristensenMentionsManagement thinker admired by Eric Ries whose endorsement of Ries's work was personally important to him.
“I bet that if I read reread all of the Clayton Christensen stuff”
▶ Watch at ~1:05:00approx.
Lee SedolMentionsmentioned
“I'm very sad that Lee Sodel uh after losing against I think I I don't know if it's alpha go or zero”
▶ Watch at ~1:10:00approx.
Chris RockMentionsComedian cited as an example of testing material before a major special.
“I don't know if it's Chris Rock, but like there's a comedian that has this joke”
▶ Watch at ~1:18:00approx.
Charles PetzoldCoinedmentioned
“code uh by uh Charles Pzled which is uh the secret language of hardware and software.”
▶ Watch at ~1:19:30approx.
Ivan IllichCoinedmentioned
“tools of conviviiality by Ivan Illich.”
▶ Watch at ~1:20:00approx.
Marcus AureliusCoinedmentioned
“I think it's a Marcus Aurelius quote I believe.”
▶ Watch at ~1:24:30approx.

Resources · 51

GoogleMentionscompany · Google
“He was a product manager at Google.”
▶ Watch at ~1:00approx.
HerokuMentionscompany
“He ran the design team at Heroku.”
▶ Watch at ~1:00approx.
NotionUsessoftware · Notion Labs
“when I joined notion we were building a lot of chat interfaces”
▶ Watch at ~3:00approx.
FigmaUsessoftware · Figma
“we were designing the chat interfaces in Figma”
▶ Watch at ~3:00approx.
Stop Drawing Dead FishRecommendsother · Bret Victor
“there's this great talk by Brett Victor uh stop drawing dead fish”
▶ Watch at ~3:30approx.
GitHubUsessoftware · GitHub
“GitHub designers before LLMs contributed to to GitHub”
▶ Watch at ~7:00approx.
CodexUsessoftware · OpenAI
“pick your favorite, they're all the same, codeex, cloud code, whatever.”
▶ Watch at ~9:00approx.
How I AIMentionspodcast · Claire Vo
“He was on uh our sister podcast, How I AI.”
▶ Watch at ~12:00approx.
AppleMentionscompany · Steve Jobs, Steve Wozniak, and Ronald Wayne
“And I say this as a huge Apple fanboy”
▶ Watch at ~18:30approx.
LinuxUsessoftware
“you could run your own Linux distribution”
▶ Watch at ~19:00approx.
Ink & SwitchMentionscompany
“the folks at Incan Switch are are obviously as sort of at the forefront”
▶ Watch at ~20:00approx.
AirbnbMentionscompany · Brian Chesky, Joe Gebbia, and Nathan Blecharczyk
“It's something that I learned also from Brian Chesky at Airbnb.”
▶ Watch at ~20:30approx.
How Buildings LearnMentionsbook · Stewart Brand
“Stuart Bran's uh sort of how buildings learn I think idea”
▶ Watch at ~23:00approx.
The Timeless Way of BuildingMentionsbook · Christopher Alexander
“Now the timeless way of building and Stuart Bran's uh sort of how buildings learn”
▶ Watch at ~23:00approx.
SalesforceMentionscompany · Salesforce
“We don't need Salesforce.”
▶ Watch at ~24:00approx.
Notion AIUsessoftware · Notion Labs, Inc.
“thanks to notionai I finally understand and use notion.”
▶ Watch at ~26:00approx.
FileMaker ProMentionssoftware
“general tools, word processor, spreadsheet, FileMaker Pro”
▶ Watch at ~26:30approx.
SlackUsessoftware · Slack Technologies
“Slack is basically the OS for anthropic. Everything runs through Slack”
▶ Watch at ~27:00approx.
WorkdayMentionssoftware · Workday, Inc.
“Workday I think is another example.”
▶ Watch at ~27:30approx.
AnthropicUsescompany
“their time is better spent building AGI than trying to build better Slack.”
▶ Watch at ~28:00approx.
MetaMentionscompany
“I know Meta got got some flack for this recently where they're trying to create a leaderboard”
▶ Watch at ~36:30approx.
OpenAIMentionscompany
“for anthropic to win OpenAI needs to lose and vice versa”
▶ Watch at ~39:30approx.
Claude CodeUsessoftware · Anthropic
“once they're in cloud code or codeex everything is fine”
▶ Watch at ~40:00approx.
ConductorUsessoftware · Orkes
“And then um I I don't know conductor is another one.”
▶ Watch at ~40:00approx.
IntercomMentionscompany · Intercom
“you see this with cursor you see this with intercom notion is uh dabbling in it as well”
▶ Watch at ~45:30approx.
CursorMentionssoftware · Anysphere
“you see this with cursor you see this with intercom”
▶ Watch at ~45:30approx.
KubernetesMentionssoftware
“Kubernetes was much more successful than than Heroku”
▶ Watch at ~46:30approx.
ChatGPTMentionssoftware · OpenAI
“the first notion assistant was actually launched before Chat GPT.”
▶ Watch at ~49:30approx.
UnixMentionssoftware · Bell Labs
“it resembles the environment that coding agents are in with Unix”
▶ Watch at ~50:30approx.
iPhoneMentionsproduct · Apple
“I don't think anyone argued when they saw the first iPhone that it's obviously good.”
▶ Watch at ~55:00approx.
DropboxMentionscompany · Drew Houston and Arash Ferdowsi
“Dropbox is a great one, right?”
▶ Watch at ~1:02:30approx.
VercelMentionscompany · Guillermo Rauch
“I have to explain it as it's the versel it's the first versel”
▶ Watch at ~1:02:30approx.
SnapchatMentionssoftware · Snap Inc.
“And Snapchat obviously just like the disappearing photo concept is so interesting.”
▶ Watch at ~1:03:30approx.
AirPodsMentionsproduct · Apple
“then you have the AirPods and like oh they connect”
▶ Watch at ~1:04:00approx.
IKEAMentionscompany
“think of how would you build IKEA like a generational company”
▶ Watch at ~1:04:00approx.
TBPNMentionspodcast
“The TBPN guys have a really good way of describing this exact concept”
▶ Watch at ~1:12:30approx.
GitHub ActionsMentionssoftware · GitHub
“GitHub actions and their uh the um uh I don't know like it's very technical”
▶ Watch at ~1:14:00approx.
True VenturesMentionscompany
“the week that we were going to get a term sheet from True Ventures”
▶ Watch at ~1:14:30approx.
ObsidianMentionssoftware · Dynalist Inc.
“We did markdown folding, all the stuff that you now have in Obsidian.”
▶ Watch at ~1:15:00approx.
Code: The Hidden Language of Computer Hardware and SoftwareRecommendsbook · Charles Petzold
“code uh by uh Charles Pzled which is uh the secret language of hardware and software.”
▶ Watch at ~1:19:30approx.
Tools for ConvivialityRecommendsbook · Ivan Illich
“I have a weird one uh which is tools of conviviiality by Ivan Illich.”
▶ Watch at ~1:20:00approx.
Project Hail MaryMentionsbook · Andy Weir
“I liked the book and I think the adaptation was was was really good.”
▶ Watch at ~1:21:30approx.
Project Hail MaryRecommendsbook · Andy Weir
“I did like uh uh Project Hail Mary uh a lot.”
▶ Watch at ~1:21:30approx.
The Handmaid's TaleRecommendstv show
“TV show I'm late to this the Handmaid's Tale.”
▶ Watch at ~1:22:00approx.
GhosttyRecommendssoftware · Mitchell Hashimoto
“It's Ghosty terminal emulator. Like most people use terrible terminals. Don't do that to yourself. Just use Ghosty.”
▶ Watch at ~1:23:00approx.
MoshiRecommendssoftware
“there is a new one for the phone called Moshi. M O S HI.”
▶ Watch at ~1:23:30approx.
CorneRecommendsproduct
“there is an open source keyboard called uh I don't even know how to say it. Corny. C O R N E”
▶ Watch at ~1:23:30approx.
Civivi pocket knifeRecommendsproduct · Civivi
“A Cevivi pocket knife, uh, which is pretty high quality”
▶ Watch at ~1:24:00approx.
XUseswebsite · X Corp.
“I am unfortunately on X or Twitter.”
▶ Watch at ~1:26:00approx.
max.devUseswebsite · Max Schoening
“max.dev. Um is I don't even know if I linked to X”
▶ Watch at ~1:26:00approx.
Lenny's PodcastUsespodcast · Lenny Rachitsky
“You can find all past episodes or learn more about the show at lennispodcast.com.”
▶ Watch at ~1:27:00approx.

Spot an error or want something removed? Request a correction or removal.

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Hot Take· 7

Hot Take08:30

Why Designers Should Code (But Not To Ship It)

Max argues the point of designers and PMs writing code isn't to land it in production — it's that coding forces you to interrogate the medium you're designing with. He'd rather have a designer who deeply understands how an agent loop works than one who can tweak UI details but doesn't grasp the underlying material.

The value of coding is that it forces you to consider the medium, not to deploy
Understanding agent loops matters more than tweaking traditional software
Coding harnesses are becoming like the operating systems of the '90s
The goal is to master the material, not become a cog in the delivery mechanism

“I actually don't care at all whether designers write code that lands in production.”
Max Schoening · 08:30

“the reason why is to to become a master of the material, not a sort of cog in the delivery mechanism for the idea.”
Max Schoening · 10:00

#design#coding#product#ai

Hot Take11:00

Agency Matters More Than Skills in the AI Era

With AI putting skills at everyone's fingertips, Max says the real differentiator is agency — the belief that the world around you is malleable and can be changed. He observes agency is not evenly distributed, and the people who understand the world is malleable will thrive while those who cling to fixed role definitions will struggle.

AI removes the 'skill issue' excuse, so agency becomes the thing that matters
Agency is not evenly distributed in the world
People who see the world as malleable will do great
Those who cling to rigid role definitions will have a much harder time

“the thing that matters is agency and I don't think agency is very evenly distributed in the world”
Max Schoening · 11:00

“cultivate agency I think that's the that's the thing”
Max Schoening · 11:30

#agency#ai#career#mindset

Hot Take24:00

The SaaS Apocalypse Is Greatly Exaggerated

Against the narrative that everyone will build their own tools and abandon SaaS, Max argues most people don't actually want to maintain the full software stack. Software is like a garden that needs tending, and the 'as a service' part you pay for is the maintenance and specialists thinking hard about the problem. He expects tools to become more general, but still delivered as a service.

Most people don't want to maintain the full stack of software
The 'as a service' value is maintenance plus specialists solving a problem
He tried rebuilding Notion in a weekend and concluded people don't want that
Tools will trend toward general-purpose (like the '90s), but still as a service

“So I think to some degree the SAS apocalypse is greatly exaggerated.”
Max Schoening · 26:30

“Software is like a garden. you need to tend to it and the the the thing you pay for in the as a service is…”
Max Schoening · 25:30

#saas#product#software#hot-take

Hot Take32:30

Society Isn't Capped by Intelligence

Max pushes back on the labs' assumption that people always want the frontier model. Using a retina display analogy — once you can't see the pixels, you don't need them smaller — he argues cognitive tasks may saturate at 'good enough,' after which speed, cost, and running locally matter more. He's far more interested in an 'exoskeleton' than a god in a box.

The labs assume people always want the smartest, frontier model
Retina analogy: past a threshold of intelligence, you want speed and modality instead
Once models are 'good enough,' you optimize for local, cheaper, faster
Society is largely not capped by intelligence

“the analogy I like to give is a retina display, which is after I can't see the pixels, I can't see the pixels. I don't…”
Max Schoening · 32:30

“I think society is largely not capped by intelligence.”
Max Schoening · 33:30

#ai#models#hot-take#future

Hot Take36:00

Token Spend Is the New Lines of Code

Max warns against turning token spend into a metric to boast about, comparing it directly to bragging about lines of code written in a day. He doesn't care how many tokens someone spends — it's not a useful metric — though he has some sympathy for why a company at Meta's scale might use a leaderboard to push people to change how they work.

Making token spend a bragging metric mirrors boasting about lines of code
The largest software projects don't have that many millions of lines of code
Token spend is not a useful metric
He has sympathy for leaderboards only as a nudge to change entrenched behavior

“I think there's a real danger in sort of making the token spend the the metric to like boast about, which is the same as…”
Max Schoening · 36:00

“it's not a metric that's useful.”
Max Schoening · 36:30

#ai#metrics#management#hot-take

Hot Take1:07:30

We Already Have UBI — It's Called Knowledge Work

Max's half-joking, half-serious hot take on universal basic income: we already have it, and it's called knowledge work. He argues what we actually need to live and be content is far less than the hierarchy of jobs we've built, and that humans will keep inventing reasons why they absolutely must stay in the loop around agents.

We already have UBI in the form of knowledge work
What we actually need to be content is a lot less than we've built
He doesn't exclude his own job from this framing
Humans keep inventing reasons they must insert themselves into the loop

“we already have universal basic income. It's called knowledge work.”
Max Schoening · 1:08:00

#ubi#ai#work#hot-take

Hot Take1:11:00

Small Group Theory: Sometimes It's Good to Be Exclusive

Max's contrarian view is that inclusivity isn't always all that great — he believes in 'small group theory,' that the world is run by group chats of eight people or fewer. Applied to product, chasing 8 billion users could upset the top 500 million who want different things. Being okay with being exclusive and building a really good product for a specific in-group can be a winning strategy.

Contrarian take: inclusivity isn't always all that great
Small group theory: the world is run by group chats of eight people or fewer
Chasing every user can upset the top of the class who want different things
TBPN's tiny-but-influential audience is cited as proof exclusivity can win

“I very much believe in small group theory like I think the world is run by group chats of eight people or fewer”
Max Schoening · 1:11:30

“I think that inclusivity isn't always all that great”
Max Schoening · 1:11:30

#contrarian#product#strategy#audience

Explainer· 2

Explainer18:00

What Malleable Software Actually Means

Max defines malleable software as software that works closer to the interest of the people who use it than the interest of the corporation that makes it. He uses a home analogy — we'd never accept a living room we couldn't rearrange — yet that's exactly the world of glued-together apps we live in, where changing an app's behavior is usually impossible.

Malleable software serves users' interests over the maker's interests
Apps glue together UI, data ownership, and behavior into an unchangeable square
We'd never accept a home we couldn't rearrange, but we accept it in software
The real question is whether you have ownership over your computing life

“Malleable software is the idea that software works closer to the interest of the people that use it than the interest of the corporation that…”
Max Schoening · 18:00

“imagine you lived in an environment where you do not get to rearrange your living room and the kitchen has to be exactly set up…”
Max Schoening · 18:30

#malleable-software#product#software#ownership

Explainer57:00

Taste Is a Virtual Machine in Your Head

Max defines taste as the ability to run a virtual machine in your head that, given an idea, predicts whether a specific in-group will like it. You decide your in-group and get good at emulating their reaction — and the only way to build it is reps with feedback, exactly like training a model via back-propagation. This is why he's skeptical that taste is humanity's last differentiator.

Taste = running a mental VM that predicts an in-group's reaction to an idea
You pick your in-group; it doesn't have to be all 8 billion people
The only way to build taste is iterations with feedback — reps
The loop resembles back-propagation, so taste may not be uniquely human

“You're able to run a virtual machine in your head where given an idea, you can predict for a certain inroup whether they're going to…”
Max Schoening · 57:30

“And to do that you just have to do reps. It's almost like training a model”
Max Schoening · 58:00

#taste#product#design#ai

Story· 1

Story12:30

Drive Notion Like It's Stolen

Max gives concrete examples of high agency at Notion: the mantra of driving the company 'like it's stolen' even as a non-founder, Brian Leven becoming the org's number one recruiter beyond his day job, and Eric Lou reshaping his own role from writing PRDs to building prototypes so he'd be worth hiring in a founder's first five.

Even non-founders can contribute in a way that exercises real agency
Brian Leven blurs engineering and design and became Notion's top recruiter
Eric Lou reshaped his role from strategy docs to Figma to building prototypes
High agency looks like changing your role to how you think it should be

“do you drive notion like it's stolen”
Max Schoening · 12:30

“I'm going to work on the skills so that you would hire me in the first five”
Max Schoening · 13:00

#agency#notion#career#teams

Q&A· 1

Q&A16:00

How to Develop Agency: Just Start Making Things

Asked how someone can build agency within themselves, Max points to the Steve Jobs realization that the world is made by people no smarter than you — and says the biggest through-line is making. Tinkering and making things (even a home-cooked meal) puts you on a treadmill of creating, and eventually people pay attention, awakening you to the idea that you can just change things.

The Steve Jobs insight: the world is made up by people no smarter than you
The biggest through-line to agency is making things
Tinkering is innately human — tool-making, cooking, creating art all count
Don't start by scheming around a bad boss — just start making things

“the biggest throughine I've found is making.”
Max Schoening · 16:30

“no, no, just start by making things.”
Max Schoening · 17:00

#agency#advice#making#mindset

Tool· 2

Tool1:19:30

Max Schoening's Book Recommendations

In the lightning round, Max recommends three books: Code by Charles Petzold (the secret language of hardware and software), Tools for Conviviality by Ivan Illich (tools that enable human autonomy versus industrial-scale tools that erode it), and Seeing Like a State (how systems create false legibility that neglects on-the-ground reality).

Code by Charles Petzold — how computers actually work, no code until chapter 27
Tools for Conviviality by Ivan Illich — tools that enable vs. erode human autonomy
Seeing Like a State — the trap of fake legibility for executives
Executives love creating false legibility because humans crave signal over noise

“it is actually surprising to me how many professionally employed programmers don't know how computers work.”
Max Schoening · 1:19:30

#books#recommendations#reading#tools

Tool1:23:00

Max Schoening's Product Picks: Ghostty, Moshi, Corne, Civivi

Max's recent product loves: Ghostty terminal emulator (most people use terrible terminals), Moshi for the phone, an open-source Corne split keyboard (whose schematics you could fully rebuild from scratch), and a Civivi pocket knife. Several picks tie back to his theme of clawing back agency and ownership over your compute life.

Ghostty terminal emulator — don't do a terrible terminal to yourself
Moshi — a not-free but well-done phone product he's exploring
Corne — an open-source split keyboard you can rebuild from schematics
Civivi pocket knife — a higher-quality physical tool

“It's Ghosty terminal emulator. Like most people use terrible terminals. Don't do that to yourself. Just use Ghosty.”
Max Schoening · 1:23:00

#products#tools#recommendations#hardware

Takeaway· 2

Takeaway28:30

The First 10% of Every Project Is Now Free

The biggest change to Max's work is that the first 10% of every project is now free — there's no point writing a PRD when you can just build the janky demo version. But he cautions that even if the first 90% is done, the last 10% is still 90% of the work. This makes it cheap to explore many paths and builds iteration into product development early.

The first 10% of every project is now essentially free
No point writing a PRD when you can build the demo version instead
Even if the first 90% is done, the last 10% is still actually 90%
You can now send off 10 agents to explore 10 different paths cheaply

“I think the first 10% of every project are now free.”
Max Schoening · 28:30

“even if you're generous and say the first 90% are now done the last 10% are still actually 90%.”
Max Schoening · 29:30

#product#ai#prototyping#workflow

Takeaway1:00:30

Great Products Have One Tiny Core That's a Superpower

Max says the truly great products all have one tiny core that is exceptionally good — GitHub's pull request, Heroku's git push, Dropbox's menu-bar sync icon, Notion's blocks. The biggest pitfall is the belief that adding one more feature will finally make the product great. That never works; the surrounding value flows from the tiny superpower at the center.

Truly great products have one tiny core that is exceptionally good
The biggest pitfall: thinking one more feature will finally make it great
Examples: GitHub's pull request, Heroku's git push, Dropbox's sync icon
Dropbox's icon was so good at syncing it doubled as an internet indicator

“if I really look at the the truly great products they all have one tiny core that is so exceptionally good.”
Max Schoening · 1:01:00

“one of the biggest pitfalls is if you get into the loop of if I just add one more thing to the product it'll be…”
Max Schoening · 1:01:00

#product#design#strategy#takeaway

Key ideas

Frameworks in this episode

Be The User Reset (Jobs-to-be-Done)

Cultivate Agency, Not Skills

Demos Not Memos

Design in the Material

Obviously Good, Then Incremental Correctness

Taste as a Trainable Model

The Tiny Core Principle

People & resources mentioned

People · 27

Resources · 51

Insights & moments

◆Hot Take· 7

Why Designers Should Code (But Not To Ship It)

Agency Matters More Than Skills in the AI Era

The SaaS Apocalypse Is Greatly Exaggerated

Society Isn't Capped by Intelligence

Token Spend Is the New Lines of Code

We Already Have UBI — It's Called Knowledge Work

Small Group Theory: Sometimes It's Good to Be Exclusive

✶Explainer· 2

What Malleable Software Actually Means

Taste Is a Virtual Machine in Your Head

❝Story· 1

Drive Notion Like It's Stolen

?Q&A· 1

How to Develop Agency: Just Start Making Things

✦Tool· 2

Max Schoening's Book Recommendations

Max Schoening's Product Picks: Ghostty, Moshi, Corne, Civivi

▲Takeaway· 2

The First 10% of Every Project Is Now Free

Great Products Have One Tiny Core That's a Superpower

Hot Take· 7

Explainer· 2

Story· 1

Q&A· 1

Tool· 2

Takeaway· 2