Over 1,300 Stripe pull requests merged each week are completely minion-produced, human-reviewed, but contain no human-written code (up from 1,000 last week).
How we built minions: https://t.co/GazfpFU6L4.
Second Minions blog post! Covering:
* strategic investments we've made for devs that have paid huge dividends as we build coding agents.
* agent specific stuff
* And a small metrics update to reflect the growth we've had in the past week 😉
https://t.co/ABuDgByOxY
We build out a tool at Stripe called Minons! It makes it super easy for Stripe engineers to kick off async agents, tied into our development infrastructure - enabling us to ship faster and with higher quality.
https://t.co/SaxCwyt0dx
very fun conversation where @gaybrick talks about Minions - a tool my team built to accelerate product development with AI agents. more detailed writeup coming soon 👀
I asked @gaybrick how @stripe thinks about build vs buy in AI:
"We are constantly thinking about new ways to empower developers.
We built something cool recently called Minions. Which uses large language models to automate a lot of the run load that’s coming to our product teams.
You click a button in a Jira ticket, and a minion will go off and fix or update the thing. You’ll see the PR, then click a button to approve. And that’s getting a ton of use internally.
For us, build vs buy is dictated by just how strategic something is to the long-term arc of the company.
One of the principles that we think about a lot at Stripe is, what are the non-speculative, forever compounding investment areas? And we think developer productivity is one of those.
You make your developers happier, more productive, that’s just going to make you a better place for engineers to work every year.
And for developer productivity, we often find that our systems are so specific that building on top of them ourselves is a lot more efficient than trying to bolt things on."
I asked @gaybrick how @stripe thinks about build vs buy in AI:
"We are constantly thinking about new ways to empower developers.
We built something cool recently called Minions. Which uses large language models to automate a lot of the run load that’s coming to our product teams.
You click a button in a Jira ticket, and a minion will go off and fix or update the thing. You’ll see the PR, then click a button to approve. And that’s getting a ton of use internally.
For us, build vs buy is dictated by just how strategic something is to the long-term arc of the company.
One of the principles that we think about a lot at Stripe is, what are the non-speculative, forever compounding investment areas? And we think developer productivity is one of those.
You make your developers happier, more productive, that’s just going to make you a better place for engineers to work every year.
And for developer productivity, we often find that our systems are so specific that building on top of them ourselves is a lot more efficient than trying to bolt things on."
@RufusPeabody ignorant question: what's the general transferability for professional bettors? like, I'd imagine anything that is modeling+being clever is open, right? finance is obvious but there are no many applications in my world for risk modeling + some hustle.
y’all actually finishing your side projects??
As @shepwalker says — the thing i love about ai coding assistants is they let me build half finished side projects at an exponential rate