Link feed
Fresh links
- How we rebuilt Next.js with AI in one week
Cloudflare recreating Next.js with a team of two and AI tooling in a week is an impressive feat in itself, but the fact that by some metrics the remake is actually better is really something.
The process started with a plan. I spent a couple of hours going back and forth with Claude in OpenCode to define the architecture: what to build, in what order, which abstractions to use. That plan became the north star. Source: How we rebuilt Next.js with AI in one week
- It’s a busy time for sci-fi, but don’t miss Aphelion | The Verge
The new adventure game is a nice counterpoint to the recent barrage of blockbusters. Source: It’s a busy time for sci-fi, but don’t miss Aphelion | The Verge
I don't game as much as I used to, but when I do, I tend to play games like this. I haven't given it a go yet, but I'm hoping it's akin to The Invincible.
- "Spider-Noir" - Authentic Black & White Trailer | Prime Video - YouTube
Next month, the web starts to unravel. Watch the new trailer for “Spider-Noir” – a live-action series starring Nicolas Cage – arriving in both Authentic Black & White and True-Hue Full Color May 27 on Prime. Source: "Spider-Noir" - Authentic Black & White Trailer | Prime Video - YouTube
Interesting this is a TV series and not a movie, baffling there's a colour option?!
- Introducing GPT-Rosalind for life sciences research | OpenAI
On average, it takes roughly 10 to 15 years to go from target discovery to regulatory approval for a new drug in the United States. Gains made at the earliest stages of discovery compound downstream in better target selection, stronger biological hypotheses and higher-quality experiments. Progress in the life sciences is constrained not only by the difficulty of the underlying science, but by the complexity of the research workflows themselves. Scientists must work across large volumes of literature, specialized databases, experimental data, and evolving hypotheses in order to generate and evaluate new ideas. These workflows are often time-intensive, fragmented, and difficult to scale. We believe advanced AI systems can help researchers move through these workflows faster—not just by making existing work more efficient, but by helping scientists explore more possibilities, surface connections that might otherwise be missed, and arrive at better hypotheses sooner. By supporting evidence synthesis, hypothesis generation, experimental planning, and other multi-step research tasks, this model is designed to help researchers accelerate the early stages of discovery. Over time, these systems could help life sciences organizations discover breakthroughs that wouldn’t otherwise be possible, with a much higher rate of success. Source: Introducing GPT-Rosalind for life sciences research | OpenAI
There are plenty of things to be concerned about in the world of AI, but there's also a lot of hope. For people like me, advances like this one could be life-changing.
- Introducing GPT-5.5
Across all three evals, GPT‑5.5 improves on GPT‑5.4’s scores while using fewer tokens. Source: Introducing GPT-5.5
I just had a conversation today about how GPT-5.4 was a noticeable step up in coding work, more inline with Claude Opus. Sounds like GPT-5.5 improves on that. Good to see real competition in this space.
- Introducing workspace agents in ChatGPT | OpenAI
Workspace agents are an evolution of GPTs. Powered by Codex, they can take on many of the tasks people already do at work—from preparing reports, to writing code, to responding to messages. They run in the cloud, so they can keep working even when you’re not. They’re also designed to be shared within an organization, so teams can build an agent once, use it together in ChatGPT or Slack, and improve it over time. Source: Introducing workspace agents in ChatGPT | OpenAI
Yes please, I’m in. These tools are improving at pace and I'm loving it!
The process of agentic delivery is a skill of its own. The description given here aligns with my own experiences.
Side note: I've been trying OpenCode (mentioned in this article) on a private project recently and found it quite good. Using DeepSeek V4 agents, I've found it to be a good, cheap alternative to the ever-increasing costs of the larger players.