Recent Posts

see all posts →

30 October 2026

'Trust but verify'; or, 'the correctness ratchet'

The name of the game in agentic development is figuring out how to clearly tell agents what you want– and then how to be sure they follow through. Lately I’ve been tinkering with an interesting technique around this that I’d like to share.1 I call it “trust but verify”2.

The basic idea is that sometimes you have things you want to “verify” but they can’t really be mechanistically checked. In that case, you let your agent do its best. But you also have it create an evidence file that contains the inputs it drew on to make that decision along with its rationale. Then you gate your CI on checking that these evidence files exist and that the inputs are accurate. Those checks are deterministic and cheap. The rationale itself is natural language and can’t be checked.

Doing this “trust but verify” technique on its own is helpful. It creates a kind of hill-climbing towards correctness that is stronger than (e.g.) tests alone. It helps to check meta-reasoning like “is the set of tests sufficient to show this requirement is true”. Documenting the input also means that if any input should change, the CI check will start to fail, and the agent will be forced to revalidate its reasoning.

What I find most interesting though is that “trust but verify” can be used to create a kind of correctness ratchet over time. The idea is that, now that the inputs + rationale are documented, you can have other agents come back and double-check that reasoning, even when the inputs have not otherwise changed. As agents naturally improve, or just due to the randomness inherent in how agents work, they will naturally find things the original rationale overlooked, and you’ll be able to address them.

read more →

14 October 2026

What agents mean for Rust (and vice versa)

Sheldon breathing into a paper bag.
Me, calmly assessing the implications of LLMs for programming language design(ers).

As someone who has made a career of programming language design, I definitely have moments of existential dread about the rise of LLMs. I started coding in 5th grade on an x861 that ran at 5MHz, had something like a 5MB hard-disk and 512K of RAM2. Computers were really, really different back then. But honestly, programming wasn’t. For all the advances in IDEs and programming languages and so forth, the basic act of entering code into a file has remained surprisingly constant3 – right up until a few years ago, when everything kinda turned on its head. That’s been exciting and empowering at times4 but also scary and disorienting. Will my skills in programming matter? What about my skills in programming language design (such as they are)? This blog post is my attempt to answer those questions.

So as not to keep you in suspense, let me just give you the conclusion. In short, yes, I believe there will still be a role for humans in programming. And yes, I think that programming language design is still relevant. And, naturally, I also happen to think Rust is a really good fit for agentic development – but that’s not to say we can’t make it a better fit. And that’s what this post is ultimately about: what makes Rust and agentic development work well, and where we could double down on those strengths.

This is a kind of “blog post companion version” of a talk I gave this week at REBASE, part of SPLASH. You might also enjoy the slides from the talk I gave at IWACO, IWACO and Rust: hidden links revealed, which was a bit more of a retrospective (no soundtrack for those, though). In any case, this turned out to be a great conference! Jack Huey presented on the Rust Vision Doc and, quite to my surprise, Steve Klabnik and Dorian Scheidt presented “LLMs as Collaborators in Language Specification and Design”, talking about their process in building out Rue. It was fun to see Steve again. On top of that, there was a ton of great research presented at IWACO in particular. I was particularly taken by the work on Reachability Types and Information Borrowing. I also really liked learning about Hylo Lang, particularly their approach to projections. Clever and cool.

read more →

10 October 2026

Why 'externalized' proofs of cyclic trait impls does not work

For this post, I wanted to talk about two different approaches to handling supertraits. I’m calling them modular proofs vs external proofs. The key idea of this post is that, if we want to have cyclic trait impls, we really need to use a modular proof strategy, where the impl establishes all supertraits hold. Previously we had considered an external strategy, where the piece of code using the impl has the obligation to prove the supertraits hold. Modular proofs always seemed better but I did not think they were workable in the past. But I have become convinced that external proofs are incompatible with Rust as designed, and hence modular proofs are really the only option1. This post dives into that reasoning, and also gives a bit of explanation of what I mean by proofs in the first place.

read more →

10 August 2026

Cylic trait implementations: motivation

Lately I’ve been thinking about cyclic trait implementations. This is a problem that I’ve been trying to understand for years and years and I finally feel like I’m geting somewhere. I’m going to try to write out a series of blog posts documenting those explorations and, hopefully, culminating in a design that could be RFC’d. In this first post, I want to talk about one of the interesting questions, what I am going to call “internal” vs “external” proofs. I know that this material can seem abstract, so I’m going to try and connect it to “real Rust” as much as possible! This particular blog post is an introduction, explaining the general problem and giving some motivation for why we care.

read more →

15 July 2026

Battery packs: Let's talk about crates, baby

Battery pack logo

This blog post describes an idea I’ve been kicking around called battery packs. Battery packs are a curated set of crates arranged around a common theme. For example, there’s a CLI battery pack that has everything you need to build a great CLI, an opinionated pack for creating a backend web service, and one for embedded development (based on the Embedded Working Group’s Awesome Rust repository). We’ve also got some smaller ones, such as the error-handling battery pack that shows how to handle errors in Rust. But this is just the beginning – a key part of the battery pack design is that anybody can create one.

read more →

9 June 2026

Only Bounds

only bounds are going to be the most impactful change to Rust that you’ve never heard of. They are currently being designed and developed by the Arm team (David Wood, Rémy Rakic, et al.) as part of the Sized Hierarchy and Scalable Vector Extension project goal. This post explores the feature and aims to answer a particular question about the design (the scope of bounds, I’ll explain). But before I dive in, I want to give a bit of context.

read more →

see all posts →