Chris Rybicki

tinker interested in distributed systems, compilers, and sometimes philosophy. eng @datadoghq, previously @awscloud @winglangio. he/him

[Chris Rybicki @rybickic]

vim is kiki, and emacs is bouba

[Chris Rybicki @rybickic]

@snwy_me I was going to suggest https://t.co/tXBPzoMSuc just bc I learned about them a few days ago, but i'm not sure if they actually provide logos or if they only come up with names 🤔

[Chris Rybicki @rybickic]

@danilop Love the format of this (code with annotations)! The explanations feel a bit AI-ish though?

For example, on a variable named "BOS", the annotation says "A special token marking where names begin and end" without ever answering what BOS stands for ("beginning of sequence")!

[Chris Rybicki @rybickic]

Gemini doesn't seem to be great at identifying prime numbers.

99,999,991 is equal to 7 x 13 x 769 x 1429

[AJ Stuyvenberg @astuyve]

NEW from Datadog: it's Lapdog!
Ever wondered what your AI agent was actually doing?

Our latest free project runs locally and traces reasoning and tool calls in Codex, Claude Code, and Pi.

You can now see what your agent is REALLY doing, live: https://t.co/3dVBozFlPx

[Chris Rybicki @rybickic]

@stevekrouse That said, I don't know how the economics work out. And if my team's system had better end to end testing / harnesses, then the amount of time needed for a human to be in the loop could have been a lot lower. (read: I want to invest more in testing)

[Chris Rybicki @rybickic]

@stevekrouse Currently working at a company with ~4k engineers and we have interns. Just finished mentoring one for four months, and I think it was productive! Some of our team's systems aren't that well tested (a lot of behavior only happens in prod), so having a human lead changes did help.

[Mitchell Hashimoto @mitchellh]

I strongly believe there are entire companies right now under heavy AI psychosis and its impossible to have rational conversations about it with them. I can't name any specific people because they include personal friends I deeply respect, but I worry about how this plays out.

I lived through the great MTBF vs MTTR (mean-time-between-failure vs. mean-time-to-recovery) reckoning of infrastructure during the transition to cloud and cloud automation. All those arguments are rearing their ugly heads again but now its... the whole software development industry (maybe the whole world, really).

It's frightening, because the psychosis folks operate under an almost absolute "MTTR is all you need" mentality: "its fine to ship bugs because the agents will fix them so quickly and at a scale humans can't do!" We learned in infrastructure that MTTR is great but you can't yeet resilient systems entirely.

The main issue is I don't even know how to bring this up to people I know personally, because bringing this topic up leads to immediately dismissals like "no no, it has full test coverage" or "bug reports are going down" or something, which just don't paint the whole picture.

We already learned this lesson once in infrastructure: you can automate yourself into a very resilient catastrophe machine. Systems can appear healthy by local metrics while globally becoming incomprehensible. Bug reports can go down while latent risk explodes. Test coverage can rise while semantic understanding falls. Changes happens so fast that nobody notices the underlying architecture decaying.

I worry.

[Patrick Collison @patrickc]

I'm interested in "trapped buildings": those that couldn't be built today (because of zoning and code changes) but also can't be substantially modified or demolished (because of historic protection rules). One of those phenomena that really makes one wonder what exactly we're trying to do.

Has anyone ever estimated what fraction of buildings in major cities fall into this category?

When I asked Claude about San Francisco, it concluded: "If forced to give a single number with a single confidence rating: roughly 100,000 buildings — about two-thirds of San Francisco's physical structures — sit in the trap as a practical matter. Confidence: moderate. The number could be 70,000 or 130,000 depending on how strictly you operationalize "can't be substantially modified.""

[Abakcus @abakcus]

Here's Looking at Euclid by Helen Friel.

Helen Friel is a paper engineer and she make beautiful things. Look at these paper sculptures of Oliver Byrne's version of Euclid's Elements. 😍

[Andrej Karpathy @karpathy]

This is the the quote I've been citing a lot recently.

[Chris Rybicki @rybickic]

New personal best for running a half marathon!

I didn't get greatest sleep and only did a handful of training runs in the weeks before - but still feeling proud of the results. Looking forward to more races!

[François Chollet @fchollet]

You cannot think your way to a perfect design. Only building and testing, over many iterations, can reveal the flaws in your mental model and provide the feedback you need to create the best design possible.

[Marc Brooker @MarcJBrooker]

Over the last couple weeks, I've been experimenting with a new way to teach folks about stability and metastable failures in distributed systems.

[Chris Rybicki @rybickic]

@netcapgirl Imagine monitoring the situation from up there

[Chris Rybicki @rybickic]

@eatonphil "At this point I don't really want to keep trying since named format arguments were a key part of the static site generator."

brutal but fair 😂

I like how clear the writing is!

[Lydia Hallie ✨ @lydiahallie]

Excited to announce Claude for Open Source ❤️

We're giving 6 months of free Claude Max 20x to open source maintainers and core contributors.

If you maintain a popular project or contribute across open source, please apply!

https://t.co/inuh0hxREA

[Chris Rybicki @rybickic]

@eatonphil subscribed!

[Phil Eaton @eatonphil]

I started a software research company

[Chris Rybicki @rybickic]

@ekzhang1 would love to read an essay on this (I also find this oddly relatable!)

[Chris Rybicki @rybickic]

@speedymachine1 @swyx @nytimes https://t.co/KysP9XPqxl