✨ Olivia Kaufman ✨

✨ Olivia Kaufman ✨

@olivia.wtf

she/her · mathemagical software wizard @ Google (ex-MSFT) way too excited about optimization, puzzles · i have too many interests ;_; playing on Hard Mode™ · 🏳️‍⚧️, pan, poly, 🐦 fursuiter, AuDHD, DID 36/f/chicago · 🔞 minors DNI · opinions my own

172 followers483 following752 posts

えっ待って、PHPの脆弱性で阿部寛までやられるんだけど!? PHPすら使ってないのに巻き添えになってるの無理すぎる😇😇😇

i've never been so glad to not be running any php stuff _yikes_

nightshade →home, for now@nightshade.technology

details leaked. this vulnerability is a full unauthenticated unconditional remote code execution exploit against any service running on a PHP version from 5.4 until latest yeah if anything deserves CVSS 11, this is it. this is CATACLYSMICALLY bad.

Me: *spends countless hours learning about the EC crypto security model* eMudhra: The private key is 4

3cats@rachel.supply

wdym your private key was literally the number 4?????

My spouse: I'm being advertised to again. Me: Oh? My spouse: I've been offered a $5 audio book called "The Concise History of the Middle East" that is 23 hours long. Me: Well, yeah. You don't think they're going all the way back to property taxes in Ur and the cradle of civilization, do you?

The reaction of the decomp scene to AI as someone who's been doing reverse engineering for a _long_ time is wild to me. There are a lot of pain points in it, and AI (as a tool) improves those _substantially_. It might have ruined a leaderboard. I don't think it ruined the craft.

View full thread

2) It may be that RE is solved by AI. Good. I enjoy crosswords, but RE was a world where every door is covered in a crossword and sometimes that's fun and sometimes it's tedious. There will always be fun puzzles. I was solving app challenges in HTS back in the early 2000s.

3) The joy in RE is mostly the result. We apply RE to accomplish something: understanding how something works, modding something, etc. There's another puzzle on the other side of RE. There always is. The puzzles will change and there will be more to do. It's OK that it's other puzzles.

The reaction of the decomp scene to AI as someone who's been doing reverse engineering for a _long_ time is wild to me. There are a lot of pain points in it, and AI (as a tool) improves those _substantially_. It might have ruined a leaderboard. I don't think it ruined the craft.

To be clear, I think there will be a new "meta" around this: 1) There may be RE that humans are better at than AI. I saw this recently with a friend who used an AI to iteratively make parts of a binary harder for an RE AI to understand, and it got stuck in places where I didn't.

My YGO banlist bingo: - Elfnote Tinia, Elfnote Lucina, and Ame no Habakiri no Mitsurugi will be Forbidden. - They'll hit some mathmechs + an FTK piece like Beelze. - There'll be a huge floodgate sweep, but they won't touch Ash or Droll. - TTOC will be Forbidden, and Limited will be soft Toon hits.

stayed way too late at work, but also managed to react really quickly to a page that would have certainly otherwise woken me up just oncall things

these are _very_ complicated systems with a lot of emergent properties. and how they work has changed a lot in a very short time span. i think anthropomorphizing language has done a great deal of harm in communicating about them.

Will Oremus@willoremus.com

feels like there is a group of smart and well-intentioned people who long ago made up their minds that AI Is Bad, swore off using it, and now get all their info about AI from each other. not saying they're wrong that ai is bad! but i think they're gradually losing touch with what it is and does

✨ Olivia Kaufman ✨ reposted

when you ask your coworker why his small PR is +17546/-15 and he hits you with the lgtm stare

ponder@ponder.ooo

lgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtmlgtm

Rush offered me an earlier appointment time if I was willing to drive to Munster, IN (the appointment is in Chicago, IL). Like, what are we even talking about?????

Someday I hope I have the opportunity to buy a drink to whomever keeps pushing the 3D graphics production quality of CSB videos. The picture on the left is from a video 18 years ago, and the picture on the right is from a video this week.

Like, they don't have to do this to push safety (the NTSB does similar stellar work, IMO, but rarely aims for this kind of quality). And just... I appreciate it so much. Thank you, whoever you are. 💜

✨ Olivia Kaufman ✨ reposted

Okay so I've built a new Atlas, this time it lets you explore the conversations on Bluesky over the past 7 days graphically. It should update every ~6 hours and help you navigate the information environment that exists on here. I've learned a LOT about what people do here today... atlas.jazco.dev

atlas.jazco.devAtlasA living map of what Bluesky is talking about: the past week's conversations, grouped into topics and regions and rebuilt every six hours.

Decks without archetypes are so soulless to me. Like yeah... maybe it's a Light and Darkness deck with a bunch of engines. But thematically this is all over the place. You want to _know_ you're playing an anime villain.

YGOPRODeck@ygoprodeck.com

Azamina Mitsurugi Dark Magician Light and Darkness Ritual / Montreal WCQ Regional Top 8 🏆 ✅ https://ygop.dk/d/735076 #yugioh #yugiohtcg

people keep using the acronym "EA" to talk about AI stuff and I have no idea what it is. Evolutionary algorithms? Idk, this isn't in the CS papers i'm reading.

✨ Olivia Kaufman ✨ reposted

i feel like the small bsky AI community here is not really EA or e/acc but it does seem to have evolved its own three distinct strains: soberminded competent commentary, demoscene manic vibe coding, and pseudo-esoteric horny AI liberationists