after a few intense years as cofounder of @icebreaker_xyz, i made the challenging decision to transition to an advisor role as it is the best thing for the company and for me at this time
icebreaker was one of the hardest and most formative things i’ve done. building a company is humbling – market, pmf, fundraising, stamina, all of it
grateful to our team, customers, supporters, investors, and collaborators. especially grateful to @web3pm, who is continuing to carry the vision of verifiable user-owned reputation forward
as for what’s next, excited to share i’m joining @raincards as staff product designer alongside a world-class team in NYC
This is me talking to my computer without making a sound.
After just a month of collecting data, our model is already approaching dictation in accuracy. We were surprised to see that it generalizes to unseen participants as well!
(1/n)
Rain went from 20 to 120 people in 12 months.
Brandon (@wallstreetpaper) asked Charles Yoo-Naut (@cnaut), Co-Founder & CTO of Rain, how they keep the team sharp:
💬"The goal isn’t to keep the culture. It’s to evolve it in a way that makes sense for where the business is."
A closer look at scoped cards.
Stablecoin cards with programmable limits.
Built for programs with employees, contractors, vendors, and AI agents.
Define exactly what each card can do before it's issued.
New Anthropic research: A global workspace in language models.
Of everything happening in your brain right now, only a tiny fraction is consciously accessible—thoughts you can describe, hold in mind, and reason with.
We found a strikingly similar divide inside Claude.
"Money should move as quickly as information does on the internet, and we have the technology to do that. So let's make it happen." Says Brian Alapatt at @raincards
Rain closed a $250 million Series C led by @ICONIQCapital to issue stablecoin-backed cards.
A lot of companies are now setting AI adoption targets and tracking usage. How many seats are active? What share of work touches a model?
But that measures the wrong thing. It judges people on inputs.
We don't have any token leaderboards at Rain. We don't set any goals based on usage.
What we set are goals that are hard enough you probably can't hit them without AI. We give people whatever budget and tools they need to get the job done, and we centralize our data so they can actually access it. Then we judge the work, not the method.
I don't care if you wrote the report with five models or by hand. I care whether the report is good and whether it was done on time.
But AI isn't a side experiment for us. We're building agentic payments infrastructure so it runs deep in our product, not just our workflows. It has allowed us to stay relatively small while setting bigger goals. That should be the whole point.
@noahzweben i am not a GitLab company but we are on Enterprise (pretty sure) and Claude is still asking me which repo every time i tag it 😐
pls no repo claude