
This is how I revamped the security of my DGX Sparks! With all the AI hacks going left and right, you cannot be paranoid enough about security. Check it out 👇


Building local AI & security tools in public

This is how I revamped the security of my DGX Sparks! With all the AI hacks going left and right, you cannot be paranoid enough about security. Check it out 👇


Choose kindness. With kindness, everybody wins. Wise words Merve 🙌
hate to break this to you if you wake up and can't choose kindness and caring for something, you're ngmi universe has a reserved place for this type of people try it for once if you haven't

Imagine having a banking blackout in a world where fiat currency has been banned. It's always been the plan.


Idk how many times I have said it but this is yet another proof, @liquidai models are amazing, small and very capable! Go check them out, get started with Local AI today!
Local models have been called many poor names. LFM3-1.7B Designed this, look how far we have come. Local is the future. @huggingface @0xSero



Connecting Four DGX Sparks is now officially supported by @NVIDIAAI without a switch 🚀
A supercomputing lab on your desk. ⚡ New at #SC26: the NVIDIA DGX Spark HPC Developer Kit, built for researchers, academics and developers. 🟢 Connect up to 4x DGX Spark systems in a ring over NVIDIA ConnectX-7 🟢 No network switch needed 🟢 Up to 512GB of memory


Built this yesterday for @Solana Breakpoint 2026: Profile Links 🚀 An app that turns the badge into your contact card. Your handle and your links with QR codes so it's easy for anyone to scan. Everything is edited on the badge itself. Nothing leaves the microSD. No phone, no server, and it works with no wifi. There is an LED kill switch in case you want to save some battery. Get it on the app store 👇👇


Moonshot open-sourced KHA: a Go SDK + CLI for Kimi Hosted Agents. This not not a model, it's a client for long-lived agent sessions in their cloud sandboxes (not a local runtime): agents, environments, append-only event logs, signed webhooks. Release under Apache-2.0.

It's only a matter of time until they start restricting consumer GPUs purchases for local inference. Own your compute!
BREAKING: The US Commerce Department announces new restrictions on remote access to advanced American AI chips and computing infrastructure. The new rules aim to prevent China and others subject to US export restrictions from circumventing existing controls by accessing American AI technology through cloud computing services. Prior to this, a loophole allowed foreign entities to access advanced US computing capabilities without physically importing restricted chips. More AI regulation is on the way.

Local AI will win. Own your compute!
Local AI will be fast, open source and for everyone. This is my promise to you. tensorfold.dev

Tomorrow evening, Tuesday 17th Nov we are gonna gather in London, near Olympia, to talk about Local AI, models, security and anything in between. DM for details ✌️
London Local AI Folks, tomorrow evening @javierbits and I are meeting for a pint near Olympia and you should come and join us! DM me if you're interested and I'll send the details 🍻

🚨 Mainnet validators: Linux kernel bug CVE-2026-99288 (NoisyTCP) requires URGENT action. Instructions apply to all Agave nodes, staked or unstaked. No validator restart needed. Patch your Linux distro. No patch yet? Disable SYN cookies as a stopgap. Steps 🧵