Post

What happened to this company is awful. There's another side to this though which I want to discuss. It relates to a story from my own past and there are lessons to be learned for the next few months and years. twitter-thread.com/t/2048103471... Back in 2008 I was doing some consulting work...

twitter-thread.comAn AI Agent Just Destroyed Our Production Data. It Confessed in Writing. by @lifeof_jer(JER) | Twitter Thread ReaderAn AI Agent Just Destroyed Our Production Data. It Confessed in Writing. A 30-hour timeline of how Cursor's agent, Railway's API, and an industry that markets AI safety faster than it ships it took d...

4 reposts20 likes

Mostly helping start ups, small businesses, local libraries. I'd been doing this as a side gig for about 6 years at that point. AWS had launched in 2006 and in 2007 I started to see a lot of my sysadmin gigs dry up. "Our developers can do all that themselves now, we don't need you."

At the time I was young and naive and figured that was that. I limped along for a few years. Then in 2010 I saw a huge surge in new business. The story was always the same: "We were wrong, it turns out reliability/scaling/production readiness isn't easy. Please help. We're in trouble."

It took 2 years to go from "We're in a new easy-button world" to "Oh my god what have we done." Jer and PocketOS were themselves just trying to run a small business and do the best they can. They trusted their infrastructure (and now development) providers. They weren't wrong to do it.

What everyone will start learning (again) is the other side of the equation: We're responsible and accountable for our work to our customers and the latest tech isn't a shortcut. It's a risk. Reliability is *challenging* on a good day. Most days it feels Sisyphean. You can't skip the humans here.