I have been finding AI really useful for coding, but this is a good reminder that AI benchmarks are not representative of reality, and developers cannot be trusted to accurately estimate their work, even after completing it 🙃

Bluesky Social

METR (@metr.org)

We ran a randomized controlled trial to see how much AI coding tools speed up experienced open-source developers. The results surprised us: Develo...

Starting a thread to coordinate a test of my social-key-backup tool with @lizsweig and @Hazey

@hzrd149 is there any way for me to report impersonators from Nostrudel?

I'm still looking for a few more folks to help me out with this! View quoted note →

I'm looking for a few people to help me test my social key backup tool for Nostr! Hoping to get at least 4 folks. I'm not going to actually back up my real key to you, I just want to run through the process with others and make sure it works for other people. You'll need python installed and some familiarity with the command line. If you'd be willing to help me out with this tomorrow please reply to this or send me a DM! And I'm considering creating a Nostr group for comms, lmk if there is a group chat tool you like.