Thinking Systems

It’s perhaps the most efficient investment by @zeddotdev. Students are the near-future engineers and entrepreneurs and humans in general tend to retain their tools of choice over their careers because of the experience and muscle memory the train.

Just tried @jetbrains Air. Looks like Jetbrains Fleet repurposed (which is not a bad idea). But there's no ACP integration (can't use claude-code) and no mermaid diagrams rendering. I'll try again when at least these two are addressed.

Respect for @AnthropicAI. They chose to stay meaningful to their own mission statement.

@sama One chicken tell the others; the fox has deep respect for our safety 🥴

Tonight, we reached an agreement with the Department of War to deploy our models in their classified network. In all of our interactions, the DoW displayed a deep respect for safety and a desire to partner to achieve the best possible outcome. AI safety and wide distribution of benefits are the core of our mission. Two of our most important safety principles are prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems. The DoW agrees with these principles, reflects them in law and policy, and we put them into our agreement. We also will build technical safeguards to ensure our models behave as they should, which the DoW also wanted. We will deploy FDEs to help with our models and to ensure their safety, we will deploy on cloud networks only. We are asking the DoW to offer these same terms to all AI companies, which in our opinion we think everyone should be willing to accept. We have expressed our strong desire to see things de-escalate away from legal and governmental actions and towards reasonable agreements. We remain committed to serve all of humanity as best we can. The world is a complicated, messy, and sometimes dangerous place.

@AnthropicAI my engineering org uses several ai agents to build software today. I'm trying to align and bring them to use claude-code teams. I think a hackathon could be a great start. Would you be interested in sponsoring a virtual hackathon for my org? Lovable has reached out.

I think likewise with the only addition that writing code with test cases and a compiler still is an important exercise. It keeps validating and course-correcting your understanding, thinking and approaches.

I broke down the entire Claude AI ecosystem in 30 minutes. Tokens, context windows, agents, system prompts, MCP, ACP, Claude Code, skills. Every component explained from first principles and connected into the full picture. youtu.be/Hc8TI39njHI?si…

Such an amazing talent, Tony. I love every single video, every second of it.

A question I keep getting from friends and colleagues: "How much should I delegate to AI agents vs. stay in control?" My take: AI changed how we program. It didn't change the engineering. My analysis in a video. The meaning & risks youtube.com/watch?v=AgZdRZ…

@zeddotdev I love Zed. Keep doing the great work. Maybe a canary release would help (with increasing percentage or rollout) based on the telemetry data on error threshold.

In this week's stable release (v0.224), we saw an uptick in app crashes around the new split diff feature (1.03% of app opens). We've addressed most of the panics via patches we've published in the last 24 hours—please update to stable v0.244.10. In the coming release cycles, we'll be dedicating engineering focus to improving Zed's stability and safeguarding the stable channel going forward.

@thomaswue @brunoborges @mariofusco @langchain4j That’s why vibe coding is only suitable for prototyping (non-engineers mostly). Not a production codebase. It should be AI generated and properly code-reviewed, each PR.

@brunoborges @mariofusco @langchain4j The class cast exception seems to be btw used to implement regular control flow. This is a VM compiler optimizers dream 😀!

What's the german word for: "being satisfied of having added a nice html visualization for @langchain4j agentic systems, but also feeling guilty because you actually vibe-coded it in 2 hours instead of hand-writing it in 2 weeks" ? github.com/langchain4j/la…

Is it possible let set @zeddotdev as a diff viewer in @claudeai cli? So far I can only integrate Intellij/Rust Rover.

@criccomini @martinkl Have you briefed anywhere what’s new in the second edition? I actually loved the first edition and think this book is one of a kind.

The 2nd edition of Designing Data-Intensive Applications, by @martinkl and me, is finished and sent to the printers! Ebooks available next week, and print books in 3–4 weeks. Sigh of relief. 😅 (BTW, this is a good opportunity to support your favourite local bookshop!)

@BenjDicken It should “Covers Rust 2026 Edition” isn’t it?

Next book stream? Maybe we can lock in Elon as a guest.

Cargo builds help me test my MacBook Pro fan every now and then 😀 Just a rust thing.

Meet a specialized AI that listens to Arabic Quran recitations and transcribes them with remarkable accuracy. This isn't just another speech model. It's fine-tuned specifically for the unique cadence and vocabulary of Quranic Arabic. Perfect for developers building Islamic tech tools.

@rustaceans_rs I tend to disagree there buddy. Vulnerabilities in C and C++ don’t happen because of skill issues when you have a team of expert engineers. But they do happen nevertheless. I think unsafe Rust code should be highly discouraged no matter the amount of experts around.

Unsafe Rust written by experts is safer than “safe” code written confidently.

@zeddotdev Nice. That’s one of the highly awaited features. Zed team never disappoints.

Zednesday 🎁

@zeddotdev Can’t fully until you provide mermaid diagrams rendering (in markdown), Jupyter notebooks and agent chat history

Join us, won’t u

Exactly. A $100 plan would make a good one I believe.