Sabitlenmiş Tweet
Nathan Baschez
55.3K posts
Nathan Baschez
@nbaschez
Founder of @lexdotpage. AI scout at @trueventures. Previously: co-founder @Every, first employee @SubstackInc. Always: @SoniaBaschez
Los Angeles, CA Katılım Kasım 2008
5.7K Takip Edilen47.8K Takipçiler
@eastdakota law of conservation of modularity / attractive profits is the real winner imo every.to/divinations/fi…
English
Prediction: Clay Christensen is best known for his theory of “Disruptive Innovation,” but the business transformations AI is forcing will eventually show that his most impactful work is his theory of “Jobs to be Done.”
English
I am desperately trying to find that meme video of Burt Reynolds (?) looking at the camera with a smug grin like "i was right"
Does anyone have that? Thank you 😅
English
This is amazing, I have my own experimental mode now in @conductor_build: Expand all tool calls
One of the weirder things I realize most people don't do that I do: I read all the thinking traces of Claude Code and it helps me a/ understand what is happening and b/ stop Claude from going down the wrong path.
I recommend you try it.
English
@RhysSullivan yes
the key is every data connection typically has a variety of "gotchas" and you need a markdown file to accumulate solutions
English
I don't think the article assumes the startup market is zero sum, the idea is that if you follow the same methods as everyone else, you end up pursuing the same ideas as everyone else, and any particular idea is probably for the most part zero sum even though the overall market is not
also, honestly, i also disagreed with some stuff in the article too, but it was really interesting for me to read because it's a question i find fascinating and almost nobody writes anything serious about it these days
English
Really? So many odd assumptions, like that the startup market is zero sum. New widespread tactics don’t just raise the standard in competitive spaces. They help founders unlock net new spaces. And more companies being founded as “founder” becomes a prestigious title means succcess rate may be lower amongst people who shouldn’t have been a founder, and higher for legit founders applying the tactics.
English
This is the most interesting thing I've read in like 3 years colossus.com/article/we-hav…
English
i probably actually agree with many of your critiques about his answers, but still very much enjoyed reading it because it is so rare for someone to even take the question seriously, and i am fascinated by the question
where i would maybe disagree is with the idea that "capitalism is working" means that "we have made progress towards a science of startups / innovation"
yes capitalism is working, no we have not made any real progress in understanding how to innovate beyond obvious platitudes
English
@nbaschez @patrick_oshag I found it kinda shallow, lots of gaps in data cited. Judge based on outcomes, capitalism is working and the quality of life available to us is better than ever. Tons of innovation and consumer choice, especially with open source.
English
if you're a pm and i send you this a day later, it means i unblocked myself, but you might not love how i did it
tweet davidson@andyreed
if you’re a pm and i send you this, it means i’m blocked from shipping
English
@bragebang sure! but also to be transparent so am i haha
still though the market is huge and this is more of a foundational paradigm than a single company thing so i would love to meet and trade notes
English
@nbaschez working on something cool for multiplayer / collab! Can I ping you when we're ready to share?
English
@dieworkwear The lean back, hands folded, absolutely diabolical
English
@snowalgo Coding agents are genuinely very different from everything you just named imo
English
@nbaschez People were connecting to LLMs through telegram, whatsapp, etc since '22/'23, Mindy had the email connection, there was autogpt & agentgpt....
English
Openclaw the paradigm is incredibly important
Openclaw the product leaves a lot to be desired imo
Justin Mitchell@jmitch
Claude Dispatch already crushing tasks that OpenClaw failed at
English
@snowalgo I think the new thing is “what if I left Claude code running, and hooked it up to a messaging app so I could text it”
English
@nbaschez there was nothing new about the paradigm, it just went ultra viral for some reason
English
@evanjconrad Either way I think Figma is a decent bet, but it logical to me that if the uncertainty goes up a lot, that the price might go down until the future feels easier to see
English
@evanjconrad I’d be curious for your take on this, but imo the key question is not if someone can cheaply clone Figma, it is whether we need design software like it at all anymore, or if designers will basically vibe code designs
English
I think people’s mistake here is thinking that software creation costs are going to zero; they’re not, they’re just going down. Inference is expensive!
when you make the cost of labor go down, this harms companies who’s product has overshot the customer expectation, and are relatively simple to get 90% of the way there. New features produce less interesting returns to value.
If that’s not the case (companies that have lots to build), then the returns of diminishing labor go to the incumbent players with lots of capital to deploy (far more than any customer might have)
evan conrad@evanjconrad
figma seems to be on sale
English
In case you’re ever in this situation: the insurance broker is not the person to ask
They don’t make money by getting insurance to pay, they make money by getting you to buy insurance
Water damage restoration contractors, on the other hand….
Nick@nickcammarata
had major pipes bursts at a property i have. insurance broker reviewed the policy and was sure it wasn't covered. a few years ago would have stopped there, but put the giant pdfs into claude, said he thinks it's covered and would handle claims process. got payout today wohoo
English
