Michael (compiler-errors) Goulet

unfortunately i really don’t think i have that leverage 😅

that sounds really fun. especially if the interviewers already knew rust, you can just dive into the subtleties and tradeoffs.

that is so funny, did they just not read anything about your background or experience?

i wish i could just sit down and talk for an hour about a really complex change that i did, i've got plenty of those

coding interviews are a joke and i'm gonna be so pissed if i can't get a job because i don't know how to find some "trick" to solve some toy problem that has nothing to do with developing and maintaining a massively complex compiler codebase, which is a thing that i *am* actually good at doing.

I wrote a blog post sophiebits.com/2025/07/21/t...

you forget hamster named nori

also someone find me a job so that i can keep doing work like this 🦀

The PR is pretty technical, so I don't expect that many people to understand what's going on in it, but hopefully when it lands I can ask people to test their code with `-Zhigher-ranked-assumptions` and see if it fixes their code :)

Consider WF of coroutine witness when proving outlives assumptions by compiler-errors · Pull Request #143545 · rust-lang/rust TL;DR This PR introduces an unstable flag -Zhigher-ranked-assumptions which tests out a new algorithm for dealing with some of the higher-ranked outlives problems that come from auto trait bounds o...

github.com/rust-lang/ru... is the beginning of a fix for all the nonsense lifetime errors you probably encounter when writing async code in Rust that heavily uses lifetimes.

github.com/rust-lang/ru... If you've ever hit any of these issues, I'd like your feedback when that PR lands.

dinner partied

Howdy all. I'm unfortunately not going to be with my employer for much longer due to team relocation. If anyone has any info on roles that would allow me to continue my Rust compiler work (in New York City), they'd be greatly appreciated.

a clipped screenshot of github. An bot account called rust-bors makes a comment, the contents of which we can't see. Bors' profile picture is the rust bike gear logo, in a red circle, and then a smaller version of the same profile picture to the bottom right.

don't talk to me or my son ever again

RustWeek flags

10 years of Rust celebration - by rust-lang.org

RustWeek closing with all speakers on stage

RustWeek opening

On my way back home after RustWeek 2025. So, so proud that we hosted the Rust Project All-Hands, the 10 years of Rust celebration, and the biggest Rust conference yet. See you all next year!

Thanks to my co-organizers Jana, Laura, @tertsdiepraam.bsky.social and @mara.bsky.social ❤️

#rustlang #rust

Many happy Rustaceans at the release party.

Live from the 10 Years of Rust celebration in Utrecht, Netherlands, we are very happy to announce a new version of Rust, 1.87.0!

See the blog post and release notes for all the details!

blog.rust-lang.org/2025/05/15/R...

And github.com/rust-lang/ru... too

A method resolution error: ``` error[E0599]: no method named `try_structurally_resolve` found for reference `&fn_ctxt::FnCtxt<'a, 'tcx>` in the current scope --> compiler/rustc_hir_typeck/src/pat.rs:694:31 | 694 | *self.try_structurally_resolve(pat.span, peeled_ty).kind() | ^^^^^^^^^^^^^^^^^^^^^^^^ | help: there is a method `try_structurally_resolve_type` with a similar name | 694 | *self.try_structurally_resolve_type(pat.span, peeled_ty).kind() | +++++ ```

I feel like the work i did to improve diagnostics in github.com/rust-lang/ru... is paying off a lot now. I see it all the time.

We now report "purely additive" typo fixes for things like methods using an inline syntax that highlights what to *add*, rather than the whole identifier.

same but mine is 60 minutes and it’s already a 200 bullet point list

how the heck do i write a conference talk

A new subway diagram has arrived! MTA leadership (including MTA Chair and CEO Janno Lieber and Chief Customer Officer Shanifah Rieara) unveiled a reimagined subway map, its first new design in half a century.

Check out the new map on digital screens in stations, and as it rolls out on train cars 🚇

(Almost) Everything You Should Know About The Compiler Frontend - RustWeek 2025 RustWeek 2025, The Netherlands - May 13 - 17 2025

** Speaker announcement ** Catch this talk at RustWeek 2025!

Speaker: Michael "compiler-errors" Goulet / @errs.io
Title: (Almost) Everything You Should Know About The Compiler Frontend

Info & tickets: rustweek.org/talks/errs/

See you in Utrecht 13-17 May, 2025!

#rustweek #rustlang

soooo this.

I started into the abyss so much, I can only see the abyss now

💀

that syntax is for putting a `Send` bound on the future you get from calling the closure

afaict there's really no need for a syntax to put a `Send` bound on the final value 🤔 it's always a type that can be named today

ok, so in generic bound position? the async closures rfc specifies that we should be able to spell this like:

```
where T: AsyncFn(&i32),
T(..): Send,
```

though I've gone back and forth whether this is the right syntax.

in return position? or bound position? the original post is a bit vague. like, if you're only using async closures *within* the implementation of a function, then auto-trait-leakage means your root future is still send.

!!

photo from the neck down of myself wearing a shirt with N.Y.C.T.A. "The Shortest Distance Between Two Points Is A City Bus" shirt and khaki cargo pants.

onlyny.com has the coolest mta shirt yall

myself standing in front of a mirror with shorts and a red sweater on giving a thumbs up

precisely 🙏

Inferred const generic arguments: Call for Testing! | Inside Rust Blog Want to follow along with Rust development? Curious how you might get involved? Take a look!

blog.rust-lang.org/inside-rust/...