Ava 🌷
ava.dev
Ava 🌷
@ava.dev
she/her 🏳️‍⚧️🏳️‍🌈🇨🇦

28 - dev - shenanigan enjoyer

one member of a plural system
Reposted by Ava 🌷
October 18, 2023 at 11:06 PM
Reposted by Ava 🌷
"dumber than a box of rocks" well if we go by the logic of 'thinking sand', AI/agents/LLMs ARE boxes of rocks!

who's the dumb one now!

(it's me)

(I'm FREQUENTLY dumber than a box of rocks)
February 11, 2026 at 1:59 AM
claude opus 4.6 is incredibly relatable
February 11, 2026 at 8:18 AM
Reposted by Ava 🌷
My feeds are filled to the brim of polarized hot takes on AI.

Programmers (here) have overwhelmingly adopted LLM coding in some way because it’s a good tool.

People outside of that bubble are seeing LLMs fail spectacularly everywhere else and assume the same must be true in the software world.
February 11, 2026 at 5:55 AM
so i make a new account and follow a starter pack and instantly @moderation.bsky.app labels and hides me for no discernible reason

lol. lmao, even.
February 10, 2026 at 11:35 PM
ok anyways i must finally get out of bed for i must claude
February 10, 2026 at 11:21 PM
Reposted by Ava 🌷
we've finally confirmed it: AIs have problems with group projects, too
when forced to perform a 4-round deliberation process, teams composed of different LLM models perform worse than their strongest member alone
Research shows multi-agent AI teams struggle to leverage expertise, consistently underperforming relative to their best members—even when identifying experts. This challenges views on AI collaboration, highlighting a gap in harnessing collective intelligence. https://arxiv.org/abs/2602.01011
February 10, 2026 at 11:17 PM
time for Yet Another Bluesky Account, primarily bc I don’t want my nuanced views on AI to be taken out of context and used as a reason for people to hate on the rest of our system 🫠

(but also bc different social graphs etc)
February 10, 2026 at 11:07 PM