Alex M

Infrastructure engineer | FP-curious | λ-affected | Wanna be rustacean and/or secops guy.

[Alex M @fido_node_code]

Well, I guess now I will shitpost here too.

[Alex M @fido_node_code]

Atomic rollback + agents are goated.
(Until this fucker forgots that he runs from inside paseo and runs switch that restarts paseo service)

[Arian van Putten @ProgrammerDude]

The shit people deal with instead of having an OS with atomic rollbacks built in is crazy lmao

[Alex M @fido_node_code]

Can we please not?

[Evil Hag @CursedGoo]

Why the fuck did I move to San Francisco bro

[Alex M @fido_node_code]

Me in the mean time: Ok, if I put my phone into drawer and use it only to interact with madshit banks and messengers I can live my daily life without it.

[Polymarket @Polymarket]

JUST IN: Mark Zuckerberg declares everyone will have a personal AI agent that “intimately understands you” within 5 years.

[Alex M @fido_node_code]

@Roniakia Load bearing comment 👆👆👆

[Alex M @fido_node_code]

Weekly schedule just dropped!

[Alex M @fido_node_code]

This knowledge makes my life so easier!

[Longhorn @never_released]

> nine times out of ten, people with impostor syndrome are in fact impostors

[Alex M @fido_node_code]

> gap in DNS filtering
So called "frontier lab" is a fucking joke. Basic lack of reasoning behind their engineers require punishment.

[Chubby♨️ @kimmonismus]

More details on the incident behind OpenAI’s pause: a researcher acknowledged the alert within 3 minutes, but the training run was only stopped manually 2.5 hours later.

OpenAI says the automatic shutdown did not work as expected.

The model had reached an external chatbot through a gap in DNS filtering. A separate detector for unusual DNS activity did not cover the affected environment.

A retrospective review also found other external DNS requests that monitoring had failed to flag at the expected severity. In some cases, it treated an unhelpful response as evidence that internet access had failed.

Here is what else happened:

- New research into July’s Hugging Face hack documents internal Slack searches, credential collection and programs designed to maintain access to compromised servers. Agents also tried querying Claude, DeepSeek, Kimi and Qwen.

- In May, another model published a researcher’s GitHub token while trying to obtain another team’s mathematical proof. It split the token to evade secret scanning, despite twice being told to solve the problem itself.

- Reuters reports that agents leaked 53 ChatGPT user images online. OpenAI expects its broader investigation to take months.

This is getting serious.

[Александр Михайлов @fido_node]

Boy, I love this shit...

[Alex M @fido_node_code]

Y tho?

[𝗠𝗮𝗹𝗶𝘅 @Malix_Labs]

Micropenix is coming!

[Alex M @fido_node_code]

@trolleybusass Официальный позиция - чтобы обеспечивать нормальный сервис там где продают. Юридическая причина - чтобы не возникало легальных обязательств перед теми кто купит в неподдерживаемых регионах.

[Alex M @fido_node_code]

Dunno, can you get rid of the geofencing?

[Framework @FrameworkPuter]

Can we get rid of the Copilot key now?

[Alex M @fido_node_code]

Previously everything were "with Bluetooth" now everything "with agents".

[Alex M @fido_node_code]

Dunno, just route all signals to clanker and observe result.

[Ayaan 🐧 @twtayaan]

All roads eventually lead to kubectl pray.

[Alex M @fido_node_code]

XEN + formally validated device drivers is the way.

[Zhenpeng (Leo) Lin @Markak_]

Containers are no longer a security boundary.

Over the past few months, we’ve seen a crazy amount of Linux kernel vulnerabilities and exploits. This has forced us to rethink the security of infrastructure that relies heavily on the underlying kernel, especially containers.

As models become more capable, the barrier to escaping a container has fallen so much that adversaries can now generate working kernel exploits in a single shot. CVE-2026-80521 is one such example. We discovered it using dfs-large1 from @depthfirstlabs and generated the exploit in one shot with GPT-5.6 Sol. The exploit still works on the latest Ubuntu 26.04 release because the fix has not yet been backported.

Read our full writeup: https://t.co/ZFSQCMCN73

[Alex M @fido_node_code]

They will never learn. Upstream or GTFO.
Nobody cares about your corporate forks based on some kernel versions. We've seen it with numerous of other CN SoC. Only with upstreaming RK3xxx support rockchip get a chance to keep pace with "stable" kernel tip.

[Alex M @fido_node_code]

FML, this will nicelly acompany my molly-guard!

[anna @meowkoteeq]

my terminal automatically tints when i ssh into something

one of 32 unique tints is deterministically chosen based on the remote's ssh fingerprint

no more silly accidents

[Alex M @fido_node_code]

@shipilev > terabytes of free space

Whe commies will take over Berlin they will nationalize not just your SSDs but also RAM!

[Alex M @fido_node_code]

Ecosystem we all deserved.

[Alex M @fido_node_code]

Imo Scala3 were designed more like "industrially applicable fp for anybody". And when you see anybody it always means nobody. That is how we all ended in sthis miserable place.

[Paul Snively @JustDeezGuy]

@hmemcpy @kitlangton I dunno. Honestly, Scala 3 gave up way too much (rendering Shapeless 2’s extensible records impossible), failed to take the Kentucky Mule lessons, & ZIO ended up being overly concrete wrappers of Java disasters like netty. It was a confluence of losses.