
Founder https://xitter.cf/t.co/wqQOFGqUWE • Open-source @meilisearch • Blogging https://xitter.cf/t.co/RNoRfonWJL • Low-key writing a novel • 🇬🇧🇫🇷🇹🇼
Taipei
Joined March 2014
- Tweets7.2K
- Following284
- Followers649
- Likes7.4K
Laurent retweeted
[BREAKING] TES ZUIAN 🇨🇳 will Miss Worlds 2026 due to visa issues. He cannot travel to the tournament after failing two visa interviews
369 🇨🇳 will join the starting lineup for now
ZUIAN and the team are actively striving to secure a third interview opportunity.
Via TES on Weibo
Nuxt 4.6 is out, alongside nuxt/cli v4 🚀
🧩 nuxt/server, our first big step to a server-agnostic Nuxt
🎯 typed $fetch was rebuilt (no more TS2589!)
🔐 appSecret + sealed cookie sessions
⚡️ Vue Vapor support
🐛 brand new dev error pages
… plus lots of groundwork for Nuxt 5 🏔️
👉 nuxt.com/blog/v4-6
Laurent retweeted
We turned ourselves into League of Legends champions with a virtual cosplay and won 1st place 🥇 at the “Open AI DevDay Exchange Community Hack Day” in Taipei~ 😎
Thank you @harryworld for hosting the event and to everyone who came to try it out and voted for us! 💕
We won a full year of ChatGPT Pro, so here‘s to a full year of tokenmaxxing 🔥🤪 ty @OpenAI
#LeagueOfLegends #Cosplay #ChatGPT #tokenmaxxing
Laurent retweeted
New trophy unlocked 🏆
@striftcodes earned 1K Clicks: Reach 1,000 Google Search clicks.
Trophy #10 in their pirate chest.
Laurent retweeted
Replying to @ThePrimeagen
I've been reflecting on this. I'm generally a very optimistic person, but lately I've been feeling a strange sense of sadness that coding has been "solved".
I can, and have, made many caveats to "solved"... but I also know that in five years none of them will matter. It will be solved.
Building things is still fun, but a different kind of fun. It doesn't activate the same parts of my brain as trad coding did.
I think it's okay to mourn the thing we loved doing being different now. Software was the first job to go through the five AI stages of grief, which means we're also some of the first to internalize that this will happen to many other jobs.
It was the best of times, it was the worst of times.
Laurent retweeted
We're adding support for AGENTS.md to Claude Code.
Starting today in version 2.1.277, if there is no CLAUDE.md in a folder, Claude will check for and use AGENTS.md.
You can toggle this behavior in /config.
Laurent retweeted
millennial midlife crisis be like
maybe i should reinstall world of warcraft – at Burbank, CA
Laurent retweeted
Traditional RAG makes your LLM a passive summarizer at the end of an assembly line.
You pick what to retrieve and when to stop.
Agentic search hands it the search tools and lets it decide.
@StriftCodes shows how to build it in TypeScript using the new Meilisearch AI SDK and the Vercel AI SDK: meilisearch.com/blog/agentic…
Laurent retweeted
in the past anyone who could program well had the potential to elevate themselves in the global job market regardless of class, location, and even resources to some extent
people joke about the "permanent underclass", it never really resonated with me but in reality it really is starting to become more about $$ and how much you can afford vs your skill as an engineer or programmer
how can you grow your skills as an (AI) engineer if you are unable to afford these tools, and does it snowball in both directions?
I feel like one of the things I want to work on are systems and programs to make this more fair and accessible
Laurent retweeted
Taipei's weekly meetup for solo builders to get more users. more money.
Format is simple: say your goal → build all day → demo it
Last week was about distribution: conversion, marketing, seo!
Video 1 shotted with antigravity with gemini flash 3.8 from raw footages, not bad!
Laurent retweeted
I've shipped support for semantic / vector search and hybrid full-text + vector search in Laravel Scout.
Works with the database (pg_vector), Meilisearch, and (new) Turbopuffer drivers.
laravel.com/framework/docs/s…
Laurent retweeted
In western culture, a young white man traditionally signals his entry into upper middle class society by making a trip to Japan
Laurent retweeted
Replying to @ledevultime
I think Cursor has the best cost/price value.
Grok 4.5+ is a bit less smart than gpt-5.4+, so you need to adapt your prompting.
But for that intelligence range, the combination of low tokens used per task and generous plan allowance makes it the most attractive imo <$100/mo

























