Thanks, @voxxed_thess, for taking this photo. My books are already outdated, but they look good here. And I look like a local rapper from Manila. 😆 😎

The first speaker for today at the #ReactSummit is the RSC Queen 👸 @aurorascharff 🔥🔥

The RSC Queen 👸 @aurorascharff together with @farisaziz12 🙌🏽 and @simeonGriggs 🔥 here at Next.js Night.

We're kicking things off with a first sneak peek at some of the incredible minds joining us at dotJS, the world's sharpest #JavaScript conference for web developers, on September 18th in Paris. Meet our first confirmed main speakers: 🟡 Karen Li – Software engineer at @github 🟡 @nicodotdev – Machine Learning Engineer at @huggingface 🟡 @LeaVerou – Web Standards Expert, CSS architect, and MIT PhD 🟡 @firt – Mobile and web developer, trainer, and author of 14 books 🟡 @devlinduldulao – Software engineer, educator, and published author 🟡 @Hmadhsan – Software Engineer Get your ticket now or your combo ticket to attend @dotaiconf, the world's brightest AI conference for AI builders on Sept 17: dotjs.io/register

Here you go: youtube.com/watch?v=HotpF1Dw…, the TanStack Query Hey API presentation I did at React Miami.

Wrote a post on building a reusable NavLink in Next.js, taking inspiration from React Router. Covers: • Render props for 𝚒𝚜𝙰𝚌𝚝𝚒𝚟𝚎 and 𝚒𝚜𝙿𝚎𝚗𝚍𝚒𝚗𝚐 • 𝚞𝚜𝚎𝙿𝚊𝚝𝚑𝚗𝚊𝚖𝚎() vs 𝚞𝚜𝚎𝚂𝚎𝚕𝚎𝚌𝚝𝚎𝚍𝙻𝚊𝚢𝚘𝚞𝚝𝚂𝚎𝚐𝚖𝚎𝚗𝚝𝚜() • Flicker-free first paint • Built-in Suspense for Cache Components aurorascharff.no/posts/build…

The RSC Queen 👸 @aurorascharff presenting now at @ReactNorway 🫶🏼

Hey, I also sleep sometimes. 🤔

I tried them all. Native gives you fewer layers between you and the error, and AI can now mirror features across two native codebases. The cryptic-error headache and abstraction tax of cross-platform just aren't worth it anymore.

Do you want to know why you still need to read the code that AI just coded for you? So you can answer Q&As during your meetings with your team members, Questions and Address Concerns, and managers.

1/ Daloy cannot automatically enforce branch protection, environment approvals, secret hygiene, runner isolation, egress policy, org settings, or deploy-platform configuration in a user’s private repo.

Wrote a post on component architecture with React Server Components. Covers: • The progression from 𝚞𝚜𝚎𝙴𝚏𝚏𝚎𝚌𝚝 to React Query to route loaders to RSCs • How Suspense shapes the loading experience • Why colocation makes components easier to move and reuse (including for agents) aurorascharff.no/posts/compo…

I built the Community Agent Template a while back, an AI Slack bot with a built-in admin panel. It's up on Vercel templates if you want a reference for building agent apps with Next.js.

Your tools could be confusing your agents and you'd never know. I've been having my agents friction log confusing moments as they build Next.js apps. • Tags every step smooth, friction, or blocked • Shows what the agent fell back on when it got stuck • Turns each friction point into a concrete fix

In this workshop, @devlinduldulao will share the workflow behind building a large-scale React app while writing only a minimal part of the codebase. Explore how strategic tooling with TanStack Query, TanStack Router, HeyAPI, and AI coding agents can unlock a new level of productivity. reactalicante.es

I noticed something about AI coding model pricing. It follows the iPhone pattern. New model drops → previous model gets cheaper → that cheaper model is still incredibly capable. You don't need the latest to ship great code. You need the right 1 for the right price. Agree😅?

Outstanding job by the whole @g2i_co crew at @ReactMiamiConf for organizing such an amazing event. I highly recommend it for learning and networking with some of the best React developers in the community. Will definitely be back next year 🫶🏼 Peace out.

We used to debate tabs vs spaces. Bad actors are now using LLMs to scan enterprise apps for vulnerabilities at scale. Maybe it's time we care more about security than code formatting. Prioritize security. Not as an afterthought — as a foundation. ✊

Copilot is moving to usage-based billing, so token usage will matter more. Caveman mode might help a little by making AI replies shorter, but the bigger savings will come from lighter models, fewer attached files, fewer agent retries, etc. Need to stop being so triggerhappy now😂