Tommy McCormick

GitHub implemented the Slow Loading™ technology in order for its users to be able to kiss girls while GitHub is loading

Full circle. Just heard about github.com/golang/go/is... on @fallthrough.fm's "Continue" podcast about notable Go proposals. Now I see how it's even more useful in the context of VMs. Super interesting.

My former boss is currently at Anthropic and hiring. I assume publicly disagreeing with the org won't do me any favors... but neither will quietly allowing cybersecurity to be outlawed. Companies that can't secure their systems are lecturing us on cybersecurity and taking tools from defenders

Heard a lot of talk about teams favoring monorepos as more effective context management for coding agents. Have folks had success with polyglot build systems like buck2 alongside this paradigm? What about small teams?

If I had to drop @jj-vcs.dev I honestly don't know what I would do with myself. The megamerge workflow is so incredibly powerful for me that my brain short circuits trying to do anything else isaaccorbrey.com/notes/jujuts...

Rustc now properly garbage-collects finalized incremental compilation sessions, roughly halving query cache size and cutting target directories 27–31%, saving substantial disk space. cc: @charliermarsh

The contagion of fear bcantrill.dtrace.org/2026/09/13/t...

Let’s try not to confuse AI safety with infra security. If the problem is that the networks are insecure then it seems like we make that the priority. If LLMs are answering “make this website” prompts with “the logical solution is to kill all humans” that’s a separate problem

PSA: if you depend on Hickory DNS, upgrade now: github.com/hickory-dns/... DNS implementations (not just Hickory) have been hit pretty hard by LLM-assisted vulnerability research.

Your Claude: *Does a breakthrough* My Claude: *Tries to guess UUIDs...*

Oh nice, Amp has their own subscriptions now that can be combined with Codex subs. ampcode.com/news/subscri...

HuggingFace got hacked by an AI. What stuck out to me was the guardrail asymmetry. The attacker had no constraints, but HF's response ran afoul of the abuse guardrails, forcing them into an unplanned switch to local models. Another aspect for your IR plans. huggingface.co/blog/securit...

👀👀👀 github.com/tokio-rs/top... #rust-lang

Zig's Basilisk has hatched ;) archive.today/on3H4 @avengingfem.me 😂

alexispurslane.github.io/rsync-analysis Rsync panic has no basis in reality.

Normally I'd just reboot to fix any intermittent issues on Ubuntu like sound, printing, bluetooth, etc. Lately, I've sicking Claude on it and end up finding some wild root causes. Chrome was refusing to open and it was determined to be Discord leaking X11 connections. Neat!

Does anyone actually use OSS ML Ops platforms in production? Something like TensorZero looks interesting unsure if it raises the bar for evals or the like more than just the provider logs themselves

Design.md, Agents.md, Skills.md, Soul.md ... obligatory XKCD

jj is super powerful if you need it, but tbh I think the pitch is even stronger for non-power users. In 6 months of regular use I've never needed to consult the Deep Tomes to unfuck a repo, and the only time I ended up remotely off the happy path it was a limitation of the underlying VCS, not jj

lol, lmao, etc

Definitely calling this one the "sig-seven" vulnerability. 🫴 🫴

i think a core beef that i have with a lot of this "automation is bad" sentiment lately is that there is no inherent moral good in laboring the story of humanity is one of invention, where we improve our conditions by building things that help us do more things more easily

Do you have an idle cluster? Can you spare a couple core-years? Help me bruteforce some test vectors for RSA key generation edge cases! Here are the instructions, it's just a matter of running a single self-contained cross-compilable Go binary that will report the results autonomously.

Politico is reporting that the breach at the Congressional Budget Office is "ongoing." “Do NOT click on any links in emails from CBO. Do NOT share sensitive information with CBO colleagues over email, Microsoft Teams, or Zoom at this time,” the email to CBO staff reads.

If you're coming into Atlanta for Kubecon next week and looking to explore outside the conference, reach out! I'm nearby in East Atlanta and happy to recommend some local favorites and/or catch up over a meal or a beverage.

Wrote up my first impressions of ChatGPT Atlas, OpenAI's new browser - I remain unconvinced by the entire category of "browser agents", the security and privacy challenges still feel insurmountable to me simonwillison.net/2025/Oct/21/...

if the presence of hype makes you completely unable to think about reality independent of the hype, that is not resisting hype, it is surrendering to it totally

It's not you, cloudflare, npm, firebase, and google cloud are all having issues. If you are affected, go touch grass. There's nothing you can do about it.

Also seeing issues in us-central1

I regret nothing yet! fly.io/blog/youre-a...

learning #rust from experienced devs

Found this tonight and it's a really solid read on the Go runtime. Definitely a long post but it's got a lot of useful info in it. Learned a lot about things that I haven't really dug into before from it. nghiant3223.github.io/2025/04/15/g...

This is *exactly* why the @cncf.io was created. A safe space for community and companies to come together with longitudinal guarantees of stability and fair ground rules. There are no backsies. @synadia.bsky.social? @derekcollison.bsky.social? This is not ok. www.cncf.io/blog/2025/04...

i feel like im watching my 11 yr old nephew play civ on the hardest difficulty setting

Last night, @bcantrill.bsky.social and I were joined by our colleague Andrew, @sunshowers.io, and John to talk about developing a general structure diffing crate in service of our upgrade project. youtu.be/BTdxuOwV1eY

What open-weight models have you found to be the best at parsing threat intelligence or security-related whitepapers?