AI coding agents are quietly eating the entire dev stack. Claude now folds and unfolds your iPhone Duo in the simulator. OpenCode rewrote its browser bridge in Rust for $60 and got 25x less memory. And researchers just found silent numerical bugs corrupting gradients in training pipelines.
The pattern? Agents aren't just writing code anymore. They're running your tools, your hardware, your math.
We're watching the IDE become the agent. That's a big deal.
|
|
In Partnership with
|
·
|
|
|
|
|
|
Today's Author
Lior Alexander. Founder of AlphaSignal and former ML Engineer. Previously built ML systems at Iguazio, Guesty, Enphase, Mila.
|
|
Top News
|
|
|
|
Anthropic's Claude Code now tests iPhone Duo foldable layouts in iOS simulator
|
|
9,727 Likes
|
|
Apple's iPhone Duo is a foldable iPhone, and building apps for it means testing multiple screen states: folded, open, half-bent. That used to mean clicking around Xcode manually. Not anymore. Claude Code now runs the iPhone Duo simulator right next to your chat. You describe what you want to test, and Claude folds and unfolds the device itself, checking how your layout holds up at each position. For existing apps, just ask Claude to prep it for Duo. It can install Apple's App Resizability skill, which scans your project for layout problems and proposes fixes. You approve, Claude applies them, then verifies on the simulator. No rewrite needed. - Requires macOS with Xcode 27.1 or later
- Install the Xcode plugin: /plugin install xcode@claude-plugins-official
- Plugin lets Claude build, run, render SwiftUI previews, and search Apple docs
This puts Swift editing, builds, simulator control, and visual checks all inside one session.
|
|
|
|
Presented by Google Cloud
|
|
Google Cloud A5X Scales to 80,000 Rubin GPUs
|
|
Google Cloud's A5X instances, powered by NVIDIA's Vera Rubin platform, scale from a single 72-GPU rack to 80,000 GPUs in one data center, and nearly a million across sites.
Now generally available, A5X links 72 Rubin GPUs via sixth-gen NVLink so each rack works as one accelerator, cooled by the system behind over a gigawatt of Google infrastructure. Built for frontier training, reinforcement learning, physical AI agents, and real-time inference.
Watch the launch video to learn more.
|
|
|
|
Top News
|
|
|
|
OpenCode ships Rust-rewritten browser tool that uses 25x less memory
|
|
6,041 Likes
|
|
OpenCode Browse just dropped v0.1.0, and it's a full rewrite in Rust. That matters because the old version was bloated: it needed Node.js, Playwright, and a pile of npm packages just to connect your AI agent to a browser. The new numbers are hard to ignore: - 25x less memory used, dropping from ~190 MB down to 5.6 MB per session
- 23x smaller binary, now just 4.3 MB instead of 53-64 MB
- No more Playwright dependency, the whole thing is self-contained
- Browser tools now exposed directly to your agent as MCP tools (think: actions your AI can call like functions)
What does this unlock? Your agent can now open a URL, scrape a list, loop over it, call your other tools, and keep going, all using your real browser with your logins already active. No headless browser setup needed. It works with Claude Code, Cursor, Codex, Zed, and any MCP client. Install via the extension or CLI.
|
|
|
|
Presented by MongoDB
|
|
Your Agent Knows Syntax. Does It Know Your Database?
|
|
AI coding agents can write MongoDB code, but they don't always follow best practices. MongoDB Agent Skills provide expert guidance on schema design, indexing, and AI retrieval, right in your coding tools.
|
|
|
|
Top News
|
|
|
|
LiquidAI ships decision models that read, locate, and extract from text, images, and audio
|
|
3,360 Likes
|
|
Liquid AI just dropped d1.2 and d1.2-ultra, two models built for one thing: making decisions from any input you throw at them, text, images, or audio. These aren't chat models. They take a question, look at your data, and return a structured answer with a confidence score. Think of it like a smart classifier you can actually trust. Five new things you can now do: - Pull a specific value from a document, like an invoice total, and get "not found" if it's missing instead of a hallucinated guess
- Find every matching passage in text, like all personal data in a note
- Locate objects in images with bounding boxes, one or many
- Estimate numbers with a confidence range
- Pick every label that fits from a list, each with its own probability
The confidence scores are actually reliable. d1.2-ultra's calibration error is 0.8, meaning when it says 80%, it's right about 80% of the time. That lets you auto-route low-confidence results to human review. Pricing: $0.02/1M tokens for d1.2, $0.042 for ultra. You only pay for inputs, zero output cost.
|
|
|
At Alpha Signal, our mission is to build a sharp, engaged community focused on AI, machine learning, and cutting-edge language models, helping over 300,000 developers stay informed and ahead. We're passionate about curating the best in AI, from top research and trending technical blogs to expert insights and tailored job opportunities. We keep you connected to the breakthroughs and discussions that matter, so you can stay in the loop without endless searching. We also work closely with partners who value the future of AI, including employers and advertisers who want to reach an audience as passionate about AI as we are.
Our partnerships are based on shared values of ethics, responsibility, and a commitment to building a better world through technology.Privacy is a priority at Alpha Signal. Our Privacy Policy clearly explains how we collect, store, and use your personal and non-personal information. By using our website, you accept these terms, which you can review on our website. This policy applies across all Alpha Signal pages, outlining your rights and how to contact us if you want to adjust the use of your information. We're based in the United States. By using our site, you agree to be governed by U.S. laws.
|
|
|
Looking to promote your company, product, service, or event to 300,000+ AI developers?
|
|
|
|
|