The win of @ABDELAESPRIELLA in Colombia is huge news for LatAm tech
3rd largest in the region (55M ppl), with entrepreneurial DNA, natural resources, and a prime location.
It has been constrained by an anti-biz gov't for the past 4 years, but that changes now. Bullish!
We're excited about the US CLARITY Act.
We think all YC companies will use crypto technology, like stablecoins, before long. Not just crypto startups, not just fintech startups, but every company.
Here's why this law is such a big deal 🧵
https://t.co/39hENfAIZk
Reminder that feature factory doesn't blow up. It hums. Everyone's busy, launches keep coming, the decks look great.
You find out it was years of motion and no movement when the renewal doesn't come, or the round doesn't close, or the reorg does.
One question pulled him out, so perhaps everyone who reads this does too.
Was on a call last week with a PM at a fintech startup who just got promoted, so I asked what he was proudest of from the last 6 months.
Long list of features. A fair bit of "you wouldn't believe the hours." Classic.
So I asked what it was all worth. Did the numbers move?
"Honestly we didn't measure. But we shipped what the clients asked for."
What rattled me wasn't his answer. It's how many sharp people in product never get taught to ask the basic questions.
Who's this for. What do they actually want. Can you tell if the thing you shipped delivered it and can you measure that qualitative or quantitatively?
That's the whole job. It's not being clever or visionary... yet, almost nobody's doing it.
I post this because I wish a post like this had been around when I was starting my career and would've saved me so many mistakes down the line.
Anthropic and OpenAI are both telling engineers to write loops.
Not prompts.
Not agents.
Loops.
That is not a coincidence.
When the two most important AI labs on the planet independently converge on the same pattern — that is a signal worth paying attention to.
Most engineers are still thinking in terms of single calls.
Input → model → output.
The engineers winning in 2026 think in cycles.
Output becomes input. The model evaluates its own work. The loop runs until the result is right.
This is the complete breakdown of what loops are, why they matter, and how to build them ↓