a working thing beats a deck
i used to sell, market and invest. with ai i can build. that changes how founders should raise: create something that moves first, then raise to speed it up.

on thursday i had a long call with a friend who builds things for a living. we went deep on how we each use ai, and somewhere in there i had a realization about fundraising that i want to share.
for most of my career i was the idea guy. sales, marketing, investing. if i wanted something built, i needed a team, and if i needed a team, i needed money. so the order was always the same. pitch, raise, then build. that order is breaking.
with the tools we have now, a person who understands a problem can produce working software, research reports and designs at a pace that would have needed a small company two years ago. my friend described his setup as having an army at his fingertips. one model plans, another executes with access to the code, and he records his design sessions by voice so the agents can file the ideas into the right project.
so think about what an investor sees. one founder walks in with a pitch deck. another walks in with a working thing. a prototype you can click, or hold, or run. even if it's rough, it's a different conversation. the second founder has shown they can create value without permission.
my advice to him, and to you, is to build first and raise to accelerate. use ai to explore the design space fast. generate a lot of options, score them against what matters, use research to narrow the opportunity. then test it in a sandbox before you spend real money on anything physical. outside capital is most useful later, for production, marketing and scale.
he added a nuance i liked. research and code are becoming table stakes. a paying customer, or some other proof that people want it, is still the strongest evidence you can bring. so a working thing beats a deck, and a customer beats a working thing.
we also talked about generalists. i've always felt a bit guilty about being a generalist in a world that rewards specialists. he made me feel better. his strength is seeing a problem across domains and putting the right pieces together. he compared a specialist to someone who sees one square on the chess board, and a generalist to someone who sees the whole board and plans a few moves ahead. ai fills in a lot of the specialist depth now, which makes seeing the board worth more.
the caveat we both agreed on is that none of this works without baseline skill. the tools amplify what you bring. someone who doesn't understand the problem will produce a lot of polished nonsense, fast.
long term, i think the best raises in this era will go to companies investors chase because something is already moving. money will follow motion more than it follows slides.
if you're getting ready to raise, here's how i think about fundraising strategy.
so what could you build this month without asking anyone for money?
the machine economy brief
one email when it matters: bitcoin, ai, robotics, and what founders should do about it. unsubscribe anytime.