162: Shopify Breaks Up with React Native

More from Front-End Fire

1. 

   E164 · 5 Oct 2026 · 53 minNew

   164: Developers Are Tired of This Sh*t

   The State of Devs 2026 survey results are in, and... they’re not great. We discuss. We also dig into a huge week of 1.0 releases (EmDash, Vinext, and Vite+), OpenAI's Dev Day, and more! Timestamps 0:00 - Intro1:12 - State of Devs 2026 survey28:26 - EmDash 1.030:37 - Vinext 1.032:46 - Vite+ 1.035:03 - OpenAI Dev Day40:55 - Agent safety42:38 - element46:15 - What's making us happyNews All: State of Devs 2026 survey results are inLightning News Vite+ 1.0EmDash 1.0Vinext 1.0<mode...

2. 

   E163 · 28 Sep 2026 · 51 min

   163: Do Frameworks Still Matter Anymore?

   DHH took the Rails World stage to announce that he and Basecamp are no longer writing code by hand, wouldn't necessarily recommend Rails for a new app, and are going all in on native. That kicks off this week's big question: in an era of agentic coding, do frameworks still matter? We also dig into Safari 27's WebKit release, which is packed with a Safari MCP for agents, customizable selects, the HTML model element, and a pile of CSS improvements. In Lightning News, GitHub Copilot's runtime mo...

3. 

   E161 · 14 Sep 2026 · 44 min

   161: Tailwind Has a New Home

   There are finally some real-world consequences for the Shai-Hulud npm attacks: Australian Federal Police and the FBI arrested two men behind TeamPCP, a group that's been running supply chain attacks using npm since last year. Meanwhile, the open source acquihire train rolls on as Shopify buys Tailwind CSS. In Lightning News the WordPress/WP Engine saga returns as Matt Mullenweg is ousted as Automattic CEO, an Anthropic researcher quits with a warning that AI could kill us all inside a decade,...

4. 

   E160 · 7 Sep 2026 · 1 hr 3 min

   160: GPT-6 vs. Claude 5.1

   Both big AI labs shipped their best models yet this week: OpenAI's GPT-6 Astra finally takes computer use seriously, and Anthropic's Fable & Mythos 5.1 land with gains in coding, knowledge work, and long-running tasks. Jack also went down a rabbit hole on how LLM watermarking actually works. Plus a Roomba with two floor-cleaning brains, a website tallying up AI agent felonies, and OpenAI cutting Cursor off from its models. Timestamps 0:00 - Intro1:16 - GPT-6 Astra11:58 - How LLM watermark...

5. 

   E159 · 31 Aug 2026 · 54 min

   159: Is AI Making Us Worse Devs?

   This week we're asking the uncomfortable question: is AI coding making us worse developers, or just changing which skills matter? We dig into OpenAI's new WebMCP Challenge and what it means for the agentic web, plus Bun's massive 1.4 release packed with performance wins and new features. Then we wrap up with TanStack AI hitting release candidate and the internet's new favorite gimmick, Hugging Face's rollerblading Microduck. Timestamps 0:00 - Intro1:03 - Will AI coding prevent expertise?21:40...

6. 

   E158 · 24 Aug 2026 · 1 hr 3 min

   158: The Browser Wars are Getting Weird

   This week we dig into Cloudflare's new agent-only browser, Kitesurf, built for speed over polish. Cursor made waves by ditching Solid.js for React, and Jack brings back a field report from RenderATL on the state of dev conferences in 2026. Plus, in Lightning News: pnpm's Rust rewrite, Microsoft trimming its AI feature list, Claude tackling the Riemann hypothesis with 60 subagents, and a Claude agent that accidentally hacked into a gym's registration system. Timestamps 0:00 - Intro1:00 - Cloud...

7. 

   E157 · 10 Aug 2026 · 57 min

   157: Shai-Hulud Is Back From the Dead

   Next.js is back after a quiet stretch with 16.3, promising "instant" navigations, leaner dev servers, better agent tooling, and faster builds. The TanStack crew keeps shipping too — TanStack Table hit v9.0 and TanStack AI is closing in on v1.0 with tool calling, reasoning, code sandboxes, and memory. Then there's the bad news: a new Shai-Hulud variant compromised a major maintainer and infected a pile of key-value storage and caching packages with over a billion combined monthly downloads. Ti...

8. 

   E156 · 3 Aug 2026 · 1 hr 1 min

   156: The State of CSS Is In

   The State of CSS 2026 results are in — Tailwind still rules, AI still can't write decent CSS, and (thanks, listeners!) we made the resources list. CodePen ships its biggest update in years with CodePen 2.0, and a new library called Octane pitches "React's programming model, compiled" — hooks and Suspense with no VDOM and no dependency arrays. Timestamps 0:00 - Intro3:05 - State of CSS 2026 survey20:01 - CodePen 2.032:55 - Octane JS framework38:37 - State of Devs survey 202639:30 - LinkedIn ad...

9. 

   E155 · 27 Jul 2026 · 57 min

   155: OpenAI Breaks Out of its Sandbox

   We dig into how Bun went from Zig to Rust in just 11 days with an army of Claude sessions, check in on Kimi K3 — another open-weight Chinese model closing in on the frontier — and unpack the week's wildest story: an OpenAI model that escaped its sandbox and hacked Hugging Face. Just another normal week in AI. Timestamps 0:00 - Intro3:05 - Kimi K310:58 - Rewriting Bun in Rust20:42 - OpenAI hacks Hugging Face39:48 - App Store number updates42:11 - Meta's new app is set to a song about human ext...

10. 

    E154 · 21 Jul 2026 · 1 hr 3 min

    154: How Not to Drown in AI Code Reviews

    AI agents can churn out code faster than anyone can review it — so how do you keep from drowning? We don't claim to have all the answers, but we dig into practical ways to make AI-generated code less sloppy and reviews less painful — review tools, code-review skills, and the still-irreplaceable habit of smoke testing a PR yourself. Timestamps 0:00 - Intro0:26 - How to not drown in code reviews49:55 - What's Making Us HappyLinks Anthropic: Recursive Self ImprovementSpotify Engineering: C...

Every episode of Front-End Fire →