Florian

@rescriptlang has been doing awesome work for years, building a joyful experience for web apps in a typed language. Jane Street can’t directly use ReScript for compatibility reasons, but I’m so excited that we get to spend some time together and share ideas with this community.

You should watch these 2 talks from AI Engineer Europe, from the "Vienna school of agentic coding": @badlogicgames "Building pi in a World of Slop" youtube.com/watch?v=RjfbvD… @mitsuhiko / @cristinaponcela "The Friction is Your Judgment" youtube.com/watch?v=_Zcw_s… They're very good.

We are live! youtube.com/live/f7rQBpQpx…

@NormalThings8 @StatsLime @scheminglunatic That's because we had both pipe-first and pipe-last operators at the same time. The latter was inherited from OCaml and only recently deprecated.

@StatsLime @scheminglunatic ReScript. It's an ML-inspired language that compiles to JavaScript. They have some design choices that piss me off, like using -> for piping instead of |>, but it's a really interesting project. rescript-lang.org

Yes, I prefer programmers choose boring technology such as Standard ML rather than something shiny like Javascript.

Surprise! We just kicked off our yearly retreat event 🚨 14 core contributors arrived in Vienna to hack on the language and ecosystem. This year's theme: "ReScript goes agentic" We want to thank our sponsors @JaneStreetGroup, CCA io and @KetryxHQ to help us make this happen! 👏

@leostera Probably Moonbit?

i've been thinking a lot lately about what my ideal programming langauge would be like now in the age of ai. you know any out there that matches this? here's my checklist: a single tool — don't need to install anything but that one tool. hell, even the language extension should be able to install the tool for me. a modern, comprehensive standard library — I want 80% of everything I want to do to be covered by the Std. Its 2026, JSON is everywhere, efficient collections are everywhere, HTTP is everywhere. A standard library should include some support for these. There can always be specialized packages out of it. a standard formatter — no questions, no knobs, just fast, ideally optimized for small diffs. an extensible linter — not just a fast one, i need a linter that i can write new lints for within my code, to help wrangle the increasing pile of slop with automated checks. agents should be able to run `tool fix` and get quick hints about how to fix up the code. ...that supports codemods! — oh yes, i want to help large refactors with little codemods across the codebase. that way agents can refactor more aggressively, faster! `tool fix` should autoapply fixes if possible an extensible LSP — not just a fast one, i need an LSP that allows packages to register new commands, and new refactors, and new hints. Something that can be tailored to the thing I'm building with ease. ...that's runnable from a CLI — and I'd like all of these capabilities to be exposed via a CLI for agents to use. Just give me `tool lsp rename-symbol ...` so agents can call it to make codebase-wide refactors procedural macros — i want to be able to make simple APIs that are verbose and descriptive and efficient, and wrap 'em up in little macros that are delightful to use. I have a feeling agents will do nightmares with this but I'm willing to give it a shot. a simple package model — i want a package model that feels intuitive and pragmatic. doesn't need to be super powerful, just give me a whole package as the unit of work. Give me common commands like `add` and `search` so the agents can find what they need quickly and get it into the project. an open package registry — i want a package registry with strong conventions about package shape, how to build it, how to generate docs, but a low-bar for publishing. Github-login might just be enough. a fast build system — i want a build system that is fast, ideally with some form of content-addressable cache that can be easily saved/shared/generationally-gc'd to make multiple agents benefit from compiled artifacts. good types™️ — (aka hindley-milner types) i know i know this is a big filter, but it's hard to give up ocaml/rust-style types once you grok the value. I'm sure I'm missing something but if you know any language that hits all these ^ let me know!

@leostera At least that should be also token-efficient

@fham_r how about ruby-like syntax with do blocks

id call it RiotML

@natedevxyz @peterpme It's using Skia under the hood.

@peterpme @fham_r why not skia?

All you need to know about building fast React native apps is in this one box

@itayad @kapilansh_twt I write my backends in frontman, but frontends only without a GUI in vi.

@kapilansh_twt I moved to frontman github.com/frontman-ai/fr…

genuine question are people still using VS Code or has everyone quietly switched to Claude Code or Codex or Cursor or am i the only one ?

Introducing react-native-nano-icons! 💎🔬 The ultimate way to handle custom icons in React Native: 🛠️ Integrated SVG-to-font build process ⚡ Blazing fast native components 🎨 Multi-color font icon support 📐 Use icons inline within RN components 🚀 Expo Go & Web support out of the box Check out our GitHub 👇 or learn more in the 🧵 github.com/software-mansi…

@WoonkiMoon Sorry...

유럽에 자주 올 일이 있지도 않은 데 하필 유로환율이 이렇게 비쌀 때라니..

@ekander_ @leostera Thanks! I starred it to inform my followers as well.

@leostera welp… github.com/lessp/arkdown-…

if i see one more markdown editor...

@mitsuhiko The global npm install. I try to have not anything installed globally on npm. For codex I use brew for instance. However it's probably a non-issue since users should be nudged to sandbox it a container anyways, which can have its own npm instance.

If something really bothers you about pi, give us feedback please :)

i made a map to monitor data centers all around the world tracks construction + nearby power plants + local AI legislation, and follows the politicians behind their bans (+ if they're getting paid to do so!)

@oxcrowx youtu.be/jEDpmj6fYiI?is…

I want to ask a question regarding ADHD. Do devs with ADHD find it difficult to program imperative languages (C, C++, Zig ,etc); but find it easier to code functional languages (OCaml, Haskell, Rust, etc)?

@onehappyfellow Or OxCaml

at this point just write ocaml

Presenting Meridian: a line to connect deterministic compute and language model AI. From Neural Turing Machines and Differentiable Transformers to The Neural Computer, there’s a rich history of trying to combine traditional deterministic computation with the wildly different architecture of Artificial Intelligence. I’ve spent the last 4 weeks creating a single neural network that has the combined capabilities of a 4B param language model and a deterministic computation engine based on Web Assembly. It allows the AI deterministic integer computations up to 2^32, control flow (while loops and if statements) and a basic filesystem - all implemented as part of the transformer neural network, no external tool calls. With this architecture adding fewer than 1 million parameters to an existing 4B param language model I can take it from <20% accuracy on arithmetic with 4-digit numbers to 100% accuracy on 4 digit numbers and 99% accuracy on arithmetic up to 2^32 without adversely affecting the language model’s performance on non-mathematical tasks. The combined model can precisely execute a range of algorithms including checking number for primeness, finding the GCD of two integers and sorting arrays.

People of pi. You can view my 18 minutes @aiDotEngineer Europe talk here. May it spark joy. And laughter. youtube.com/live/_zdroS0Hc…