@magicseth It's a regression to like p91.
Which is why most people are so impressed.
And some people are mildly worried, especially if they need p95+ in their domain.
@0xPaulius@convex Sorry about this.
We're working with AWS to get lambda updated in a more timely fashion. The Convex runtime we patched with minutes.
It's frustrating enough that we're exploring moving off lambda for the Node runtime and just rolling our own generalized runtime.
150% agree with this, and I said so publicly before in conferences, and here too.
The truth of the matter is we, as an industry, are painfully aware of the many ways in which we do things that suck. We know the issues exist but we bank on the fact that the economic incentives to exploit them are not there.
Well, now they are.
Run, do not walk, from any cloud provider who is throwing public shade at Railway (or any other provider). The better your security team is, the more internal pressure you have not to mouth off about security in Q4 2026.
They're back! @convex founders @jamwt & @jamesacowling for Realtime Realtalk live stream AMA at 11 AM PT
Live on X and YouTube
https://t.co/npQ5rZD82s
Unfortunately the industry has not yet learned how to develop junior engineers in this new era, since the senior engineers "learned by doing" as juniors, and that isn't happening anymore.
I think this is an urgent talent issue the software industry will need to figure out, because the senior engineers will retire and right now there's limited investment into their successors.
@bertie_ai Again, you would not pay $27k on Convex. You would be on a business/enterprise plan and caching cost benefits would drop your bill down to $4k to $6k.
Granted, it's still not $9, but please ask your AI to do the math accurately next time. ๐