Marco Ieni

Marco Ieni

@ieni.dev

Infrastructure Engineer at @rustfoundation.org πŸŽ™οΈ Podcast host https://ieni.dev/rustship πŸ¦€ Maintainer @release-plz.dev 🐘 Mastodon @MarcoIeni@hachyderm.io 🌐 Website https://ieni.dev

619 followers172 following212 posts

Pinned

All #rustconf talks are out now! πŸ₯³ Here's mine if you want to learn more about the CI of the Rust compiler: www.youtube.com/watch?v=Gzk4...

youtube.comMarco Ieni: "How We Made the Rust CI 75% Cheaper" | RustConf 2025YouTube video by Rust Foundation
RustConf@rustconf.com

It's been a month since #rustconf 2025, but the buzz is still strong! πŸ¦€ Revisit the magic via these videos on the @rustfoundation.org YouTube channel, incl. sessions, keynotes, & this highlight reel: www.youtube.com/watch?v=Hw_I... +Don’t miss these event pics πŸ“Έ www.flickr.com/photos/rustf...

I merged a release PR after waking up and found out now that it bumped release-plz from 0.3.174 to 0.4.0. This was supposed to be 0.3.175 as there are no breaking changes. πŸ˜… I will keep releasing the action under 0.5.x.

I'll change the release-plz logic so that this issue can't happen anymore. I want to design software that it's safe to operate at 7am. Dogfooding always helps!

this was fun

Marco Ieni@ieni.dev

1. Ask AI to open a draft PR to add benchmarks for your projects, covering the most common use cases. 2. Don't review that PR. 3. Ask AI to spawn subagents to use these benchmarks to identify up to 10 performance wins that are easy to review and open a PR for each of them.

1. Ask AI to open a draft PR to add benchmarks for your projects, covering the most common use cases. 2. Don't review that PR. 3. Ask AI to spawn subagents to use these benchmarks to identify up to 10 performance wins that are easy to review and open a PR for each of them.

Bonus things to add in point 3: * In the PR body, justify the performance improvement by reporting the benchmark results before and after the PR. * Fell free to install and use profilers to identify the major bottlenecks.

Marco Ieni reposted

Excited to chat with Orhun on the livestream tomorrow! I'll bring my bag of spicy hot takes 🌢️ It's gonna be great, hope to see you all there.

I'm excited to announce a new livestream! πŸ“’πŸ₯³ Tomorrow, I'm chatting with @predr.ag - Frontier Systems Engineer OpenAI, @rustfoundation.org board member and creator of cargo-semver-checks! πŸ¦€ We'll talk Rust performance and safety at scale! ▢️ 16 UTC: www.youtube.com/watch?v=6jDV... #rustlang

I’m now team leader of the Rust Infrastructure team! I’ll keep shipping, but I’ll also pay more attention to the load, health, and overall direction of the team.

I changed the Pull Request template of release-plz. I realized I don't care if you used AI to write code, or how you used AI, or which model you used. I just want to make sure you reviewed the code before submitting it. 🧡

View full thread

This is the most sustainable way I found to keep maintaining release-plz in my spare time. It seems like it's working. In the last month: - 39 PRs merged in main excluding automations such as renovate. - closed 18 issues. Lots of bug fixes!

And finally, a reminder to all open source maintainers working on their spare time: you owe nothing to your users! Your time is a gift you are doing to the open source community. People can't demand more gifts or more expensive one.

I changed the Pull Request template of release-plz. I realized I don't care if you used AI to write code, or how you used AI, or which model you used. I just want to make sure you reviewed the code before submitting it. 🧡

View full thread

That said, I still require your human review before submitting a PR to avoid getting tens of low-quality PRs and wasting token reviewing them. I also require human review (ideally human voice) in issues, discussions, comments, etc.

Another thing I'm experimenting with: I created a bot account (marcoienibot). Agents are free to use it to create PRs with it when I ask to fix an issue. I'm also experimenting with "review -> fix" loops. Still needs improvements because sometimes PRs becomes too big.

I changed the Pull Request template of release-plz. I realized I don't care if you used AI to write code, or how you used AI, or which model you used. I just want to make sure you reviewed the code before submitting it. 🧡

Generally, Astra will review your code first. If it says there's a better implementation I'll never review your PR. Instead, I'll let Astra + Fable 5.1 write a new PR. If Astra think it makes sense, Astra will include your commits in the new PR to acknowledge your work.

Almost all PRs I received in the last months contained AI generated code. I trust my prompts/harness/skills more then the ones of random external contributors. Why should I spend my time reviewing their AI generated code instead of mine?

TIL you can checkout pull requests locally in one click with the VSCode "GitHub Pull Requests" extension.

I haven't had enough energy to work on release-plz recently. I preferred to spend my spare time on new side projects. However, I have some trips/PTO coming up in the following two months, so I'll try to catch up with PRs and issues. Let's see 🀞

Marco Ieni reposted

We rewrote one internal tool to use Rust (instead of TS/Node) as a proof of concept/trial balloon, and yesterday switched over one of our hourly jobs to use it instead of the TS version that it had been using. Over 10Γ— better memory usage and 10–100Γ— improvement in how fast it runs. 🀩

I like how Zed integrated Agents in their IDE. It has the multi-project view (which VSCode lacks) and LSP support (which ChatGPT, Amp, T3Code and others lack). If they integrate GitHub PR reviews I will try to use it instead of ChatGPT App + VSCode!

I'm building a ChatGPT Desktop Elgato Stream Deck plugin. Here is the progress I made with the icons.

Marco Ieni reposted

Glad to see the Rust Foundation continuing to grow the ways it gives money to support Rust's development.

The Rust Foundation@rustfoundation.org

Announcing the 1st cohort of Rust Maintainers in Residence! Thanks to contributions from Google, AWS, & OpenAI, the Rust Foundation and Rust Project have raised $350K to fund 6 contributors on Rustup, Clippy, compiler, stdlib, & rustdoc πŸ¦€ Read announcement: rustfoundation.org/media/rust-p...

I still got it 😎 I have no issues with vibe coding side projects, but for critical systems like the crates-io CDN, I still want to refine the code generated by AI.

Could I have achieved the same by prompting more? Probably, but sometimes when you have an intuition it's faster to edit the code manually.

Marco Ieni reposted

Voice dictation with FluidVoice is more accurate if I use a good quality microphone. Overall, dictation stresses my arms less, and I feel like I can dictate faster than I can type. πŸ‘ I still review the results of the dictation and fix them by typing where necessary.

Here is what the #rustlang Infrastructure Team worked on in the last three months and what we want to do next!

The Rust Foundation@rustfoundation.org

Rust Project Infra Team Q2 recap is out πŸ¦€ - Bors checks: 30min β†’ 1min - YubiKeys for critical access - New RISC-V + powerpc64 CI runners - Secret scanning across rust-lang repos Great writeup from Rust Foundation Infra Engineer, Ubiratan Soares: blog.rust-lang.org/inside-rust/...