iiinova — where your digital self keeps evolving

Combine that w manually spot checking some code and automated review and we’re good.

000

The strategy is that this is reviewable and we’ll land it. We don’t need thorough human review of every line of code: instead we are focusing on verifying the test setup. If it’s tested properly (every single compiler pass produces identical intermediate state), that gives us a ton of confidence..

210

That first port was more of an exercise in “how much quality can I get for a nearly automated port” to which the answer was “nowhere near enough”. So I went back and spent a lot more time getting it right. Good experience learning the limits of agents atm.

0200

Joe Savona @en-js.bsky.social · 22/03/2026

Eagle eyed observers will notice the first commit on this branch was from *after* my post here. I threw everything out and started all over again! The original port was pretty good but had some fundamental issues. Things i had told the agent to avoid, but where it gradually drifted from the guidance

1100

“These are advanced features that can be skipped for now” Citation needed

070

Joe Savona @en-js.bsky.social · 12/03/2026

We primarily expect to publish integrations but happy to discuss on a GitHub issue once the rust version is out there and you can try it.

010

I’m thinking we’ll switch to the Rust version only, as soon as it’s ready

060

No, I think AI changed the tradeoffs

General thought is that I wish I’d had the time to port by hand. You learn by writing. But alas. Anyway yeah, will try to share more learnings somewhere

050

Once the port completes we have to iterate, test, look at perf, and then ship. Will try to get a ln experimental preview out asap.

190

There are a few design choices the agent made along the way that I want to change, but nothing major enough to stop the process. I got the key design choices locked in on the initial plan. We’ll just do some polish passes at the end.

160

One big unlock (h/t @ricky.fm) was to have the agent create custom agents for planning, implementation, review, and orchestration of the project. That’s now running for a few days with minimal intervention.

3110

We’re pretty close to having an initial Rust port of React Compiler. My approach was to iterate on the plan document (write it manually, have the agent expand, edit, repeat). Then try to run it, see what goes wrong, revert back to the plan, tweak the plan, repeat.

49812

Your post was totally fair! I appreciate you sharing your experience :-)

020

The parts of the post describing subtle nuances are about which specific ways of breaking immutability the compiler does/doesn’t catch. But developers don’t need to remember which way of crafting an error will be caught or not! They just need to remember the rule, which is straightforward.

140

Thanks for sharing your experience! We’ll take a look to see if there are any of the cases you flagged where we can detect an error automatically.

Joe Savona @en-js.bsky.social · 18/01/2026

There are two aspects: the rule & how we enforce. The rule is simple: don’t mutate stuff that crosses JSX/hook boundaries. You almost certainly follow this ~all the time, else you would have crazy bugs. What feels and is more complex is that we can’t statically catch all violations of this rule.

120

What if we all got together and agreed to pay a bit of money to the government every year so that they could put together a group of trained people to protect us from masked thugs in the streets?

250

Joe Savona @en-js.bsky.social · 01/01/2026

We’re exploring a different direction but I don’t have a link handy. Cc @capt.dev

120

Joe Savona @en-js.bsky.social · 01/01/2026

Still something we have on our list to potentially optimize further in a future release

100

Joe Savona @en-js.bsky.social · 01/01/2026

In this example each element will “re-render” but what that means has changed: since each item will largely no-op if its inputs haven’t changed. With React Compiler a component function executing doesn’t necessarily mean much work happens.

100

Joe Savona @en-js.bsky.social · 31/12/2025

I hope we’ve earned enough trust that people don’t think we’re so naive as to _only_ rely on feedback that we hear about directly. We also look at things like adoption which accounts for people bouncing off for various reasons, as well as on our own judgment/experience.

020

Joe Savona @en-js.bsky.social · 31/12/2025

Did you already have ESLint running and our plugin slowed it down, or was the slowdown from switching from a fast linter back to ESLint?

Joe Savona @en-js.bsky.social · 30/12/2025

..and avoid rewriting the core logic. But more importantly: we are very actively expanding and building upon the compiler. We’ll hopefully be able to prioritize build performance sometime in 2026 but we’ll see how things go. The more we hear about it the easier it is to prioritize so send feedback.

192

Joe Savona @en-js.bsky.social · 30/12/2025

We get that reintroducing Babel in an otherwise Rust-based toolchain slows down builds. It’s a compromise. Unfortunately the Rust-based JS tooling ecosystem is fragmented - SWC and OXC - with different parsers and AST formats. We’re still hoping to be able compile with Static Hermes

1100

Joe Savona @en-js.bsky.social · 30/12/2025

Expanding on Dan’s answer. First, everyone on the React Compiler team had significant experience in Rust or C++, and several of us had also worked on rewriting Relay Compiler from JS->Rust before that was cool. The decision to use JS was purely about practicality..

140

Joe Savona @en-js.bsky.social · 18/12/2025

You may also be interested in react.dev/reference/re... to fail the build on errors

react.dev

panicThreshold – React

The library for web and native user interfaces

Joe Savona @en-js.bsky.social · 18/12/2025

So you should generally spread in the rules from the preset, and then override/add.

100

Joe Savona @en-js.bsky.social · 18/12/2025

Great post, and it’s great that the compiler is working so well for you! One thing to be aware of w the way you set up your rules: you’re missing out on most of the validations!!! In ESLint if you specify any specific rules, it disables everything else from that preset..

210

Joe Savona @en-js.bsky.social · 07/12/2025

Recent favorite: “test plan: yarn jest path/to/feature-test.js, all tests pass”. WE DONT USE JEST and the named test file doesn’t exist

050

Joe Savona @en-js.bsky.social · 06/12/2025

Trend I’m seeing getting worse recently: a legit GitHub issue gets filed. Somehow people latch onto it as something they can spam us on. Multiple similar PRs, often from different people sharing bits of code, that look 90% plausible but with blatantly obvious tells.

3180

Joe Savona @en-js.bsky.social · 04/12/2025

Try `/terminal-setup` in Claude, that fixed it for me

000

Joe Savona @en-js.bsky.social · 02/12/2025

Have you tried using React Compiler?

010

Joe Savona @en-js.bsky.social · 01/12/2025

By default we don’t error in the Babel plugin because breaking builds is generally not preferred. But if you want to fail the build on errors, you can: react.dev/reference/re...

react.dev

panicThreshold – React

The library for web and native user interfaces

111

Joe Savona @en-js.bsky.social · 01/12/2025

React Compiler tries to automatically skip code that breaks React’s rules. Use the ESLint plugin to see the errors react.dev/learn/react-...

react.dev

Debugging and Troubleshooting – React

The library for web and native user interfaces

120

Joe Savona @en-js.bsky.social · 10/11/2025

I would offer to fix it but we didn’t build that extension :-) An official React Compiler IDE is in the works, for now we recommend using the ESLint plugin in parallel w the compiler.