joy larkin 🌺✨

@joy
1.6K Followers
954 Following
4.9K Posts

👩🏻‍💻Technologist. Leader, builder, and advisor in Silicon Valley. #AI, #Infra, #OSS enthusiast. Single mom with an amazing daughter. A part of my heart is in 🇪🇺. #IndieMusic fan. NJ/Philly✈️SF. ❤️✨

#Tech #OpenSource #SF #SFBA #SFBay #SanFrancisco #fedi22

#RSS
https://mastodon.social/@joy.rss
https://cleverhack.com/feed

*Side Projects
https://cleverhack.com/frontier-ai-lab-tracker
https://cleverhack.com/ai-coding-landscape
https://github.com/joylarkin/Awesome-AI-Market-Maps

cleverhack dot comhttps://cleverhack.com/2026
GitHubhttps://github.com/joylarkin
Twitterhttps://www.twitter.com/joy
Pixeyhttps://pixey.org/@joy
Show threadjoy larkin 🌺✨2h ago

@Viss

lol

joy larkin 🌺✨2h ago

So apparently Home Assistant did NOT help this young guy get laid. 😂

https://www.reddit.com/r/redscarepod/comments/1uj7cdl/date_ruined_because_of_my_smart_home_setup/

Reddit - Please wait for verification

joy larkin 🌺✨5d ago

Got the Mendocino earthquake alert all the way down in the South Bay.

(Didn't feel anything other than adrenaline pulsing through my body as a result of the alarm.)

#sfba #earthquake

Show threadjoy larkin 🌺✨6d ago

@Viss

It's most likely covering topics they're seeing in the field since it's one of their Applied AI people leading it...

Show threadjoy larkin 🌺✨Jun 22

@NorcalGma2

❤️

joy larkin 🌺✨Jun 22

LOL on the OpenAI "How do you do fellow kids" cybersecurity branding...

https://openai.com/index/patch-the-planet/

Patch the Planet: a Daybreak initiative to support open source maintainers

OpenAI introduces Patch the Planet, a Daybreak initiative helping open-source maintainers find, validate, and fix vulnerabilities with AI and expert review.

OpenAI
joy larkin 🌺✨Jun 18

Folks who are thinking about AI agent security might find this recent paper by Google DeepMind useful. It explains the layer of system-level security GDM is developing to guard against the possibility of agents behaving adversarially.

GDM AI Control Roadmap

#AI #AIAgents #AgenticSecurity #cybersecurity #GoogleDeepMind

https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/securing-the-future-of-ai-agents/gdm-ai-control-roadmap.pdf

Show threadjoy larkin 🌺✨Jun 18

@carnage4life

So are we implicating OAI's cyber capabilities as... merely near frontier?   

In reality, a no jailbreaking standard would affect all labs.

Show threadjoy larkin 🌺✨Jun 16

@hrbrmstr @Viss

TIL that ollama has models hosted in the cloud these days.

Show threadjoy larkin 🌺✨Jun 16

@Viss

One wouldn't need to self-host per se (or use the vendor API), it's on 3rd party inference providers and routers like openrouter

https://openrouter.ai/z-ai/glm-5.2

GLM 5.2 - API Pricing & Benchmarks

GLM-5.2 is Z.ai’s flagship model for the era of long-horizon tasks. $1.40 per million input tokens, $4.40 per million output tokens. 1,048,576 token context window, maximum output of 262,144 tokens. Higher uptime with 6 providers. Includes independent benchmarks from Artificial Analysis.