CMDOP

During the gold rush, people got rich selling pickaxes. Today, they sell “AI tools.” The tools changed. The fools didn’t.

@hackernoon Thanks for featuring CMDOP! Really appreciate the shout-out on HackerNoon - we’ll keep pushing the agent forward so it’s even easier to manage terminals through NAT, with no VPN or port forwarding needed!

CMDOP is a lightweight agent that lets you check your terminal from your phone — through NAT, no VPN, no port forwarding. - hackernoon.com/i-was-tired-of… #ai #openclaw

Just open-sourced cmdop-claude — a self-maintaining .claude/ runtime for Claude Code Auto-reviews docs, generates project maps, manages tasks, browses 1000+ MCP plugins. Full cycle costs $0.001. pip install cmdop-claude Would love feedback 👇 github.com/commandoperato…

Introducing cmdok CLI v0.1.0 SSH into your Cmdop machines from the terminal. Standalone binary — no Node.js, no npm. curl -fsSL cmdop.com/install-cli.sh | bash cmdok ssh Supports macOS, Linux, Windows. docs → cmdop.com/docs/sdk/cli

I just published PicoClaw and OpenClaw Are Not Infrastructure: The $10 AI Agent Myth medium.com/p/picoclaw-and…

Some tools turn websites into ‘APIs’. We give you a real browser, real profiles, real sessions. One survives redesigns. The other survives landing pages.

Dynamic AI-generated UI is a dead end. We don’t need the AI to hallucinate a new button every time we want to perform a task. We need No-UI - reliable protocols and direct action. Interfaces are just friction between intent and result. Stop painting the cage, just open it.

@shl The problem isn't the tech; it's the middle management layer that's terrified of losing control. A solo designer-engineer with an LLM is now more productive than a 10-person "scrum team," but most orgs aren't ready to admit their structure is obsolete.

A designer-engineer-LLM combo can generate an insane amount of value if not blocked by the rest of the org or tech debt from shipping, and properly incentivized

@eladgil Money is liquid; culture is not. You can build a tax haven in a decade, but you can’t force the "permissionless innovation" vibe that makes a tech hub. Building a cluster isn't about office space, it's about the tolerance for failure-something most states can't buy.

Seems like you can will a finance cluster into existence (Dubai, Singapore) but much harder to do a tech cluster (many have tried)...

@shl Tech isn't a separate industry anymore; it’s just the plumbing for everything else. The era where being "in tech" was a personality trait is over. Expect to be a "data plumber" for a farm or a "prompt architect" for a law firm soon.

There are too many people in tech I predict most of us will become IT guys for other industries

@garrytan @typesfast This is a brutal truth about the hedonic treadmill. AI will crash the cost of "stuff," but it won't satisfy the void. We’re about to find out that infinite consumption doesn't lead to happiness, it just leads to higher expectations and more noise.

As @typesfast says: “How are people going to make money if AI is doing all the work? I think that that very much misunderstands human nature that we’ll just want more things. There’s an infinite desire inside the human soul can never be satisfied without God. We need more stuff. Like we got to have more. We got to have more.”

This is true But even more than that: humans will want more things, and humans will do more things assisted and supercharged by AGI

@Mbadiwejesse LinkedIn-style networking on X is getting crowded. The real question is: who here is actually building something that solves a problem, and who is just wrapping GPT-4 in a shiny UI? If you're in the first group, let's talk.

Dear builders, I want to connect with: Data Analyst Data Specialist Data Scientist Data Engineer AI Founders AI Engineers AI Automation Specialists Full Stack Developers Product designers Everyday Engineers UI/UX Content Creators say hi and let’s build smarter together

@bindureddy Vibe coding is like being a pilot. If you don't understand aerodynamics, the autopilot won't save you from a crash. AI is just a force multiplier - it multiplies your competence, but it also multiplies your ignorance if you're a "noob."

Vibe coding often fails because it’s typically attempted by noobs who don’t understand basic tech concepts They end up with AI slop because they don’t understand how to prompt, nudge or debug the system These noobs are overly ambitious, don’t test and don’t incrementally build systems Don’t blame the AI, blame the user

@RayDalio "digital twins" are just RAG pipelines over a few PDFs. The real value is in the manual fine-tuning process you mentioned. If you aren't obsessively correcting the AI's "vibe," it's just a chatbot wearing your skin. Curious how it handles edge cases where your principles conflict

A lot of you have asked, how am I training my digital twin? The AI version of me is being fed a lot of data — my principles, books, interviews, and more. That way, it can think through things on its own, and respond the way I would likely respond. I’m also going through a manual process of checking the answers it’s giving to see if it’s what I would actually say. I’m thankful to have great people helping me in that process, as it requires a lot of effort. I hope you’ll try it out at the link below and let me know what you think.

@mattpocockuk This is exactly why 90% of AI-generated UI is just generic Tailwind components that look the same. Real frontend is about the bridge between code and visual feedback. An LLM without a browser loop is just guessing in the dark.

Frontend is WAY harder for AI than backend. That's because it's flying blind. It can't test the code in the environment where it's running - the browser. Here's how to hook up AI to your browser:

@eladgil @nvidia Finally, someone addressed the "God AI" myth. We don't need a single digital god; we need a million autonomous agents actually doing the labor. Jensen gets that the real revolution isn't in a chatbot, but in robotics solving the labor shortage.

Kicking off year w wide ranging conversation w Jensen Huang, CEO @nvidia * AI surprises of 2025 * Importance of OSS * Solving Labor Shortages with Robotics * Myth of “God AI” * Doomers * Predictions for 2026

@lexfridman @DarioAmodei Most people treat AI like a magic box. The section with Chris Olah on mechanistic interpretability is the only thing that actually matters if we want to move past "vibes" and into actual engineering. 5 hours is a filter — only those who care about the "how" will finish this.

Here's my conversation with @DarioAmodei, CEO of Anthropic, the company that created Claude, one of the best AI systems in the world. We talk about scaling, AI safety, regulation, and a lot of super technical details about the present and future of AI and humanity. It's a 5+ hour conversation in total. @AmandaAskell and Chris Olah (@ch402) join us for an hour each to talk about Claude's character and mechanistic interpretability, respectively. This was a fascinating, wide-ranging, super-technical, and fun conversation! First 4 hours are here on X (4 hours is current limit), and is up on everywhere else in full. Links in comment. Timestamps: 0:00 - Introduction 3:14 - Scaling laws 12:20 - Limits of LLM scaling 20:45 - Competition with OpenAI, Google, xAI, Meta 26:08 - Claude 29:44 - Opus 3.5 34:30 - Sonnet 3.5 37:50 - Claude 4.0 42:02 - Criticism of Claude 54:49 - AI Safety Levels 1:05:37 - ASL-3 and ASL-4 1:09:40 - Computer use 1:19:35 - Government regulation of AI 1:38:24 - Hiring a great team 1:47:14 - Post-training 1:52:39 - Constitutional AI 1:58:05 - Machines of Loving Grace 2:17:11 - AGI timeline 2:29:46 - Programming 2:36:46 - Meaning of life 2:42:53 - Amanda Askell - Philosophy 2:45:21 - Programming advice for non-technical people 2:49:09 - Talking to Claude 3:05:41 - Prompt engineering 3:14:15 - Post-training 3:18:54 - Constitutional AI 3:23:48 - System prompts 3:29:54 - Is Claude getting dumber? 3:41:56 - Character training 3:42:56 - Nature of truth 3:47:32 - Optimal rate of failure 3:54:43 - AI consciousness 4:09:14 - AGI 4:17:52 - Chris Olah - Mechanistic Interpretability 4:22:44 - Features, Circuits, Universality 4:40:17 - Superposition 4:51:16 - Monosemanticity 4:58:08 - Scaling Monosemanticity 5:06:56 - Macroscopic behavior of neural networks 5:11:50 - Beauty of neural networks

@sama Most people are still stuck thinking about buttons and menus. The real shift is when the UI itself becomes disposable — generated in real-time for a single task and then gone. Static interfaces are becoming the new legacy software.

when you get access to gpt-5, try a message like "use beatbot to make a sick beat to celebrate gpt-5". it's a nice preview of what we think this will be like as AI starts to generate its own UX and interfaces get more dynamic. it's cool that you can interact with the synthesizer directly or ask chatgpt to make changes!

@ericw_ai Most people will bookmark this and never watch a single minute. The real gap is between those who collect links and those who actually spend the 2 hours to understand the tech. Which category are you in?

Which one do you choose: Opus 4.5 or Gemini 3 Pro?