A friend told me today: “We have a colleague named Bob. Ask him anything, and he just copy your question to an AI agent and sends the answer back.” Every company has a Bob.

Jev, now open source: Lev A 4B open source System One model based on Qwen backbone The best performance for it's small size huggingface.co/interfaze-ai/…

wait.. @eBay does not have dark mode for their website?? no way..

he’s wired in

so "Decision model" is the new name for "zero-shot classification"?

We did not create an Interfaze shitcoin. Thanks 👍

Everything that Vercel does is a banger. Straight up for the devs.

We use @vercel for a lot more than our frontend at Interfaze. We’re serving 50 million requests a month while managing GPUs without a separate backend infrastructure I’m joining @MadsMcIlwain live to walk through our setup and share why we built it this way Sign up below

We are now live on the @merge_api Gateway! Gateway lets developers access the best models across multiple providers through a single API. You can now use Interfaze through Merge for deterministic tasks where accuracy and consistency matter. Think vision based tasks like OCR and object detection, web based tasks like search or scraping, and audio based tasks like speech-to-text or speaker diarization. Happy building! Docs: docs.merge.dev/merge-gateway…

things in sf feel increasingly absurd. i find myself taking increasingly drastic measures to maintain sanity. to drown out the noise, to talk to people with good heads on their shoulders, to spend time around people who care about other people.

We've come a long way from MNIST.

Our first hardware project coming real soon

All you guys complaining about astra should be made to use opus 5 on max. this is what it writes for prod: fml

▲ ~/𝚛𝚊𝚞𝚌𝚑𝚐/𝚘𝚜𝚜-𝚐𝚛𝚊𝚗𝚝𝚜/ 𝚕𝚜▐ (v2) I'm announcing v2 of my open source grants of $1,000USD to 35 awesome contributors. The themes: ① Agent skills & tools Skills are now a valuable form of software, like Emil's world-class design eng work. Projects like 𝚙𝚐𝚋𝚘𝚝, 𝚔𝚗𝚒𝚙, 𝚞𝚗𝚕𝚒𝚐𝚑𝚝𝚑𝚘𝚞𝚜𝚎 create amazing AI engineering loops. ② Local AI Projects like 𝚜𝚕𝚘𝚝𝚜𝚝𝚛𝚎𝚊𝚖, 𝙻𝚘𝚌𝚊𝚕𝙰𝙸, and 𝚌𝚘𝚕𝚒𝚋𝚛𝚒 are making inference not just a demo but for many a real and growing utility. Exciting to watch local AI unfold. ③ Performance There's no excuse for software to be slow anymore. Projects like 𝚏𝚏𝚏 and 𝚝𝚊𝚔𝚞𝚖𝚒 leading with speed, compilers like 𝙿𝚎𝚛𝚛𝚢𝚃𝚂 and 𝚙𝚘𝚛𝚏𝚏𝚘𝚛 that can help us upgrade entire ecosystems to native, and 𝚐𝚙𝚞𝚒𝚡 pushing the GPU rendering speed frontier. ④ High quality foundations AI is still at its best when it can reuse great components than regenerating every time from first principles. Lots of love to folks working on @shadcn, Svelte(kit), Nuxt, Node.js and security foundations. ⑤ Cool experiments I couldn't resist donating to gloom.sh, a 'Finance terminal in the terminal', or beautiful funky things like 𝚍𝚒𝚝𝚑𝚎𝚛-𝚔𝚒𝚝. 🔗 rauchg-oss-grants.vercel.app… PS: These are personal, not affiliated with Vercel, and fully unconditional. If you are listed, I'll be DM'ing you or using your public contact info!

Sorry, physicists, but AI can’t test your untestable theories.

I've been making launch videos like these and want to make more professional ones (and I have a lot to learn) If there are storytellers/ creators who're into motion graphics, storytelling for tech, podcasts etc. I'd love to get you coffee and learn about your processes.

We have tons of GPUs right now and we need them to go brrrrrrrr

Introducing DefaultModel Where we only host 1 model, the best open source model We're launching with GLM 5.3 Flash The community votes and we host it GPUs are limited so we want to focus resources on the best model Consistent throughput and high uptime

BIG ANNOUNCEMENT FROM HUGGING FACE TODAY: We're unveiling Microduck 🐥🤖 It's a tiny $399 open-source robot you can teach new tricks with reinforcement learning. It can walk, pick things up, get back up when it falls, and even roller-skate. Welcome to the era of open-source affordable robots to democratize physical AI and world models! 🤗🤗🤗