Loris Cro ⚡

VP of Community @ Zig Software Foundation • Zig Livecoding • Creator of • Host of • 📧 [email protected]

[Loris Cro ⚡ @croloris]

Asynchrony is not Concurrency

https://t.co/BhE9cVq7XU

[Loris Cro ⚡ @croloris]

Sometimes I run out of steam, sometimes I go down the wrong path, and some other times I deliberately take detours, but nobody can trick me into mistaking lesser stars for my true destination.

https://t.co/mynHilpF7i

[Joran Dirk Greef @jorandirkgreef]

Andrew’s interview with JetBrains on investing in human contributors passes 400,000 views (and not stopping).

[Loris Cro ⚡ @croloris]

@gordon_cassie I have debug-only asserts in my programs as well, but beware that a dbgPanic function will not guarantee that the expression passed to it will be elided in non-debug builds, for that you need to explicitly guard the assert with an if statement.

[Loris Cro ⚡ @croloris]

You Must Fix Your Asserts

https://t.co/TCwtqsKQNP

[Loris Cro ⚡ @croloris]

@theodorebeers Also so much time is passing that the comparison between last Zig Bun and what they have now makes less sense by the day.

But I'm sure that small details like that won't stop marketing from using Jarred's name to make the wildest claim.

[Loris Cro ⚡ @croloris]

Ah so this is the blog post we were promised, a paragraph inside of an Antrhopic product announcement. Good shit.

https://t.co/VMEvjVvfDU

[Loris Cro ⚡ @croloris]

@theodorebeers You're probably right, but at this point it's not Jarred's post anymore, it's going to be Anthropic's marketing follow up to the product announcement, no matter how they dress it.

[Loris Cro ⚡ @croloris]

@rvragul_ whoops, fixed!

[Loris Cro ⚡ @croloris]

New Zig Devlog about ELF linker improvements, featuring 30ms rebuilds of a Tetris game.

Incremental is coming and y'all not ready.

https://t.co/n51gYLrDjY

[Loris Cro ⚡ @croloris]

For some people AI is an autoadaptive dunning-kruger system.

[Loris Cro ⚡ @croloris]

Mitchell is the kind of person you would love to have sitting next to you at a Zig Day

[Mitchell Hashimoto @mitchellh]

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.

As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!

I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.

It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?

88ms => 1.5ms
150K allocs => ~500 allocs

Incredible right? Nope.

My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.

This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.

The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.

Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

[Mitchell Hashimoto @mitchellh]

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.

As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!

I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.

It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?

88ms => 1.5ms
150K allocs => ~500 allocs

Incredible right? Nope.

My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.

This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.

The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.

Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

[Loris Cro ⚡ @croloris]

The next Zig Day Seattle (WA) is going to be on Jun 13!

Make sure to RSVP if you plan to attend!

https://t.co/snHoXAmvVe

[Loris Cro ⚡ @croloris]

About LLMs at Zig Days

https://t.co/uj4m5vp52u

[Joran Dirk Greef @jorandirkgreef]

Andrew is relentlessly building the world’s most powerful systems programming language (and toolchain).

He has my full support and trust (and frankly, sheer gratitude for holding firm on quality).

Props to @jetbrains for putting together a terrific interview.

[Loris Cro ⚡ @croloris]

Andrew's interview with JetBrains

https://t.co/7SsmfjksJP

[Loris Cro ⚡ @croloris]

well it's been fun rolling in the mud, I'm logging off to do actually relevant stuff, see you all next drama

[Joran Dirk Greef @jorandirkgreef]

People often get this wrong:

But per systems thinking, correctness is not a { language, static allocation, testing, etc. } problem.

Correctness is a design problem.

[Mitchell Hashimoto @mitchellh]

Fork your dependencies, trim them to only your use case, never update unless it breaks for your users. I’ve been vocal about this for 10+ years. I’ve always said that updating is way riskier than latent bugs (which can be tracked and CVEs monitored).

If you are updating a dependency, it’s on you to analyze every single commit in the full transitive set of dependencies. If you dont see anything compelling, dont update!

I remember at HashiCorp once in awhile an engineer would try to update a dep or replace a DIY lib with an external one and id always ask “show me the commit we need.” Dont update for the sake of it.

Feeling pretty swell about this mentality with all the supply chain attacks happening.