mastodon 4.7.2+glitch.techlgbtClippy team member at #Rust, "Optimizing Clippy & Linting" Project Goal owner twice. Google OSPB Award winner in 2023, featured in maintaine.rs and Rust Bytes newsletter.
Gave talks at RustChinaConf2025, Kangrejos 2025, Rust In Paris 2026 & RustWeek 2026
Not a gymrat anymore, now into contact sports (currently doing Muay Thai)
Usually worried about software performance and human rights.
Profile picture drawn by Ryuuseee, https://www.pixiv.net/en/users/270369
I'm mapping ALL crates.io and their dependencies.
These are just 14341 crates (palindrome number and about 5.9% of crates.io) and it already seems something of cosmical proportions. Like some sort of megastructure. And we have some data!
Out of 14341 crates, 46% depend directly on tokio, 4.8% depend directly on syn, only 44 depend on a rustc_* internal crate, almost all of them being rustc_version and even being the most downloaded crate of all crates.io, only 166 depend on `hashbrown`.
Reckless, a Rust-written chess engine is currently battling Stockfish in the TCEC superfinal. This means that Reckless, the only Rust-based chess engine to reach this level, has beaten every other opponent thus far.
Another 765 million instructions optimized in Clippy. If you want to see me talking about performance, I'll be at Rust In Paris IN JUST 3 DAYS.
Antoher 3.3% shaved off of Clippy :partyparrot: 
Finally got around to self-hosting a #Forgejo instance. Also finally got around to creating systemd services for all my services.
No one will complain every again that I forgot to boot something after the lights go out!
I'm currently on interviews with maintainers of:
- Lua
- Sqlite
- OpenSSH
- Clojure
- And more!
Because a future post/video(?)/talk will be released about these interviews. Hoping that you enjoy the insight these people have to bring.
THE CLIPPY CHANGELOG CAT CONTEST 1.94 IS OPEN SUBMIT YOURS NOW 💥 💥 💥 💥 💥 💥 💥
Currently working on optimizing a Clippy lint that takes 320 million instructions, turns out that 18% of all Clippy is spent in the same function.
Yes, Clippy spends 18% of the Clippy-exclusive time on the same lint.
Because of my Spanish blood, today I landed at RUST WEEK 2026 COME GET YOUR TICKET, and found myself surprised (again, this happens often) with the idea that the rest of the world eats before 10 PM.
The “dinner” time here starts at 6 PM, absolutely ridiculous!
Making it easier to sponsor Rust contributors.
> If you ever wanted to support the development of Rust "in the small", it is now simpler than ever.
https://blog.rust-lang.org/2025/12/08/making-it-easier-to-sponsor-rust-contributors/
Just 9,5 hours before taking the train for the bus in order to get to the plain straight to RUST IN PARIS 2026, another 20% performance optimization is delievered straight to your door.
I know that I make a big mess out of "performance without removing features" but this one does hide a feature. `clippy::nonminimal_bool` will be allow-by-default at least until we figure out a way to simplify it and stop it from using scary math.
320 million instructions STRAIGHT to the fire. Once again, performance wins.
I've optimized `clippy::manual_is_ascii_check` by 97.125% in the `wasmi` crate. This makes a `cargo clippy` run in `wasmi` take 820 million instructions less, approximately 3.3% of the total analysis runtime.
The main driver is a rapid change in how software is being built. Since the second half of December 2025, agentic development workflows have accelerated sharply
Since December 2025 I haven't noticed an uprise in code quality or tool diversity. But I've noticed that I've received more spam pings than usual.
The pull request establishing a t-funding Rust team has entered Final Comment Period!
Talking with @bal4e@tech.lgbt in Rust In Paris 2026 I'm now thinking again about something she said: "People don't realize how fast today's CPUs are, and they don't realize how fast CPUs made 20 years ago also are".
So, today I'm noticing how slow Github is, and I have a 3.8-4.7 GHz CPU. And it's still insanely slow.
How can some software be this slow?
Currently commissioning some Samurai stickers for my #Rust Week 2026 talk on "Redesigning the Incremental System". Make sure to be there, only 7 days left. :dragncool: :dragncool: :dragncool:
My talk at Rust In Paris will have a live demo section, and preparing for this live demo section I decided to test some popular crates and their performance improvements over time ^•ﻌ•^ฅ♡
These were the results! Make sure to check out the full talk once I've given it! See you in Paris
Rust Clippy announces that it will require age verification software for users residing in California
https://blog.goose.love/posts/rust-clippy-will-require-age-verification/
Do you want your art commissions to help a kitty? Do you like human art (tm)? Would you like a 50% DISCOUNT on all human art (including #furry, chibi and #nsfw)?
My girlfriend is offering a 50% DISCOUNT on ALL art commissions, go check her out.
More information here: http://blog.goose.love/posts/commission-art-to-help-alastor/
Boosts greatly appreciated, all proceedings go to the kitty's surgery. If you don't like Instagram but would still like to make a commission, get in contact with me and I'll figure things out.
@fasterthanlime@hachyderm.io This happened to me when trying to create a pretty-print for floats following IEEE 754 :/
Today I made my longest bash command.
```bash
for dir in ./*/; do cd $dir && for f in ./callgrind.185master.*; do if [ "$(callgrind_annotate $f | rg "clippy-driver" --stats | rg "(\d+) matches" | rg "\d+" --only-matching)" -gt "200" ]; then echo "$f - $(callgrind_annotate $f | rg 'PROGRAM TOTALS' &)"; fi; done | tail -n 1 >> difference.md && for f in ./callgrind.out.*; do if [ "$(callgrind_annotate $f | rg "clippy-driver" --stats | rg "(\d+) matches" | rg "\d+" --only-matching)" -gt "200" ]; then echo "$f - $(callgrind_annotate $f | rg 'PROGRAM TOTALS' &)"; fi; done | tail -n 1 >> difference.md; done
```
