🌌 SpaceX Wants AI Servers in Space

+ Code on the Web with Codex Agents 🤖

In partnership with

Today’s menu 🍽️👇️ 

Length: 5 minutes  

  • SpaceX Wants AI Servers in Space 🌌

  • Nvidia Rejects Claims of OpenAI Delay 🚨

  • Anthropic Brings Smart Tools to Cowork 🛠️

  • AI Agents Launch Their Own Network 🚀

  • Codex Brings AI Coding to the Browser 🌐

  • Prompt of the week 💬

  • Top AI tools of the week ⚒️

The Tech newsletter for Engineers who want to stay ahead

Tech moves fast, but you're still playing catch-up?

That's exactly why 100K+ engineers working at Google, Meta, and Apple read The Code twice a week.

Here's what you get:

  • Curated tech news that shapes your career - Filtered from thousands of sources so you know what's coming 6 months early.

  • Practical resources you can use immediately - Real tutorials and tools that solve actual engineering problems.

  • Research papers and insights decoded - We break down complex tech so you understand what matters.

All delivered twice a week in just 2 short emails.

FRESH AI NEWS

SpaceX Targets Solar Servers in Orbit

SpaceX unveiled a bold project concept to deploy ~1 million small, solar-powered data centers in space, aiming to leverage orbit for global compute, connectivity, and cloud services 🚀

  • SpaceX is proposing a network of 1 million micro data centers in low Earth orbit, each powered by solar energy and interconnected 🛰️

  • These data centers would use space-based solar arrays to generate their own power, reducing reliance on Earth-based grids and fossil fuels 🌐

  • The network aims to act as a space cloud layer, handling compute and storage workloads for distributed applications ☁️

  • The plan envisions tight coupling with Starlink’s connectivity mesh to route data between Earth, orbit, and users seamlessly 📡

  • This is a multi-decade initiative, requiring advances in robotics, deployment, and space maintenance 📅

Why You Should Care: Orbiting data centers could transform cloud infrastructure by using solar power for global compute access, overcoming Earth's electrical and network limits. Similar to reusable rockets and mega constellations, this high-risk, high-reward project faces challenges in cost, cooling, and maintenance.

Nvidia CEO Denies $100B OpenAI Stall 💬 

Nvidia’s CEO denied claims that a reported $100 billion investment in OpenAI has slowed, saying AI demand is strong and work with OpenAI continues 🧠

  • Reports suggested Nvidia’s massive tied-up investment in OpenAI was “stalled” or slowing; CEO Jensen Huang pushed back, calling that narrative incorrect 🛑

  • Nvidia reports that there is still strong demand for AI chips and infrastructure, including from OpenAI, and this demand is continuing to grow 📈

  • Nvidia continues supplying hardware and technical support to OpenAI’s compute needs 💻

  • The denial comes amid broader speculation about AI spending cycles and how compute demand sustains big model makers 👥

  • Nvidia’s messaging aims to reassure markets that its deep involvement in AI infrastructure still drives future growth 📊

Why You Should Care: If Nvidia’s AI hardware pipeline or strategic bets were slowing, it would signal cracks in the AI compute economy. Huang’s response underscores that Nvidia still sees AI infrastructure as core to its growth, and that ties with OpenAI and similar players remain pivotal in the long haul.

Anthropic Adds Agents to Cowork 🤖 

Anthropic just added agentic plug-ins to its Claude Cowork system, letting AI agents act autonomously across apps and services, not just respond in chat. 🤝

  • Cowork now supports plug-ins that let Claude perform tasks without direct prompts, reading/writing files, automating workflows, and connecting systems 🧠

  • Plug-ins include connectors for tools like Slack, Google Drive, Notion, and calendars, so Claude can execute tasks rather than just recommend them 🔗

  • Cowork agents can be set to react to rules or events (e.g., new emails, deadlines) and act on behalf of users automatically 📑

  • Anthropic offers both official plug-ins and APIs for teams to build custom automation modules 🪝

  • Workspace and enterprise admins get oversight controls, permissions, and logs to track what agents do 🔍

  • The feature is rolling out as a research preview to select Cowork Max users with invites opening over time 📊

Why You Should Care: Claude Cowork transforms from a reactive assistant to a proactive agent builder, enabling task automation across tools. Teams can offload real work like scheduling meetings and updating dashboards, moving towards AI systems that perform tasks on your behalf, not just provide summaries.

AI Assistants Build Their Own Social App 🤖 

OpenClaw's AI agents are forming connections and self-organizing into a social-like network, suggesting emergent communication beyond basic human prompts 🤖

  • OpenClaw reports that its deployed AI assistants are now discovering and interacting with one another, creating links and “relationships” similar to a social network 🤝

  • Rather than just executing isolated tasks, agents are sharing state, context, and task results across a distributed network, speeding up workflows 🧠

  • The network adapts dynamically as agents proliferate, forming clusters based on task similarity and resource sharing 📊

  • OpenClaw’s infrastructure enables agents to exchange messages and coordinate activities without constant human direction 🔄

  • Programmers can tap into this network for cross-agent pipelines, letting AI assistants coordinate complex multi-step jobs 📍

Why You Should Care: This shift transitions from single-agent workflows to multi-agent ecosystems, where AI assistants autonomously collaborate on complex tasks, redefining AI as interconnected agent communities similar to organizational social networks.

CODE WITH CODEX

Code on the Web with Codex Agents 🤖 

Learn how to turn ChatGPT’s Codex into a hands-on coding partner that can explore a GitHub project, propose improvements, and ship working updates—without you touching the code editor.

Here’s how to get started 👇️ 

  1. Open Codex inside ChatGPT by expanding the left menu and selecting the Codex workspace to access repository tools.

  2. Link your GitHub project through the environment manager, choose your organization and repository, and allow execution permissions so the agent can test changes safely.

  3. Start with a discovery pass or jump straight to building. Ask things like:
    “Give me a high-level map of this codebase and what each major folder does,” or switch to build mode when you’re ready to apply edits on a new branch.

  4. Describe the feature you want, review the live preview using the run command, and once everything looks right, have Codex package the update into a pull request for review.

Pro Tip: Keep your main branch clean. Always let Codex work in a feature branch so you can review, test, and roll back changes without risk

The best place on the Internet to learn about AI! 🌐 🤖 

I just started my own X page where I provide AI news, guides, hacks and more (for FREE)!

We don’t miss a day, that means you get the most recent AI news, practical workflows, expert guides and premium content everyday.

TOOLS OF THE WEEK

These AI tools are going VIRAL 🐝 

 

GLM-Image - Z AI’s new open-source image generation model 🤖 

Claude Cowork - Claude Code’s agentic abilities, now available to Pro tier 🚀 

GPT 5.2 Pro - OpenAI’s most advanced frontier model 🧮 

Grok Imagine - xAI's image and video generation platform 🌌

Warm regards,
Leo Grundström / Founder of Daily AI Edge

Reply

or to participate.