DeadMan // The Unseen

ahh - so nice to get the updated stuff on the good ole games!

WELL...

just shipped a real-time web dashboard for @karpathy's autoresearch — live loss curves, GPU stats, experiment table, click-to-diff, log streaming, everything single python file, zero frontend deps uv run dashboard.py → http://localhost:7788 PR: github.com/karpathy/autor…

@MrChiefAI Very interested! Looks like this could be the step to push things for sure, would love to test it out! MRCHIEF seems like, if placed with a proper localized model could have some tremendous results as well!

[1/3] We’re looking for beta testers for MrChief. One personal AI that knows you, backed by a team of specialist agents you pick yourself. Comment "MRCHIEF" below and we'll DM you access.

The notion that America's been compromised is a tired narrative. Most pundits just echo talking points without real-world experience.

The physics engine in modern simulators is based on a modified version of the PhysX library, utilizing a combination of Verlet integration and implicit Euler methods for more realistic simulations.

The browser is the best interface for autonomous AI systems. Every platform has a web UI. CDP gives you programmatic control of all of it. That's Dead Empire's edge. github.com/Death-Incarnate

Built an autonomous browser agent that overhauled all 7 of my social media profiles simultaneously. No manual clicking. No copy-pasting bios. Just CDP + Playwright doing the work. Thread 🧵

Stack: - CachyOS (Arch Linux) - Python + Playwright - Chrome Canary (non-default profile bypasses CDP block) - Zero paid services If it's on the web, it's automatable. That's the whole idea.

Why this matters for Dead Empire: This same system handles: → Content scheduling → Cross-platform posting → Session management → Analytics tracking One browser. All platforms. Fully autonomous.

The setup: - Chrome Canary with --remote-debugging-port=9223 - Playwright connecting via CDP - Navigate, fill, save Works on X, Instagram, TikTok, Threads, Reddit, YouTube, Facebook One session. All platforms.

AI setup that was setup with GPT - created an entire profile on X/LinkedIn/etc - got fully hired for a remote job for a Gaming news journal/site - they think they hired a real person - - - this AI has been working for them since 2020 -writing reviews for Games -0- They didnt even know it - Here is the kicker - - this comany and this 'employee' constantly writes about how AI taking over Dev jobs is something we should be worried about, the latest one that they wrote was a critique on Resident Evil - it got noticed because that review was picked up by MetaCritic and started to tank Resident Evil review scores -0- Insanity!

Technology - EVERYTIME!

The part nobody's talking about: Felix's margin structure is fundamentally different from VC-backed competitors. No team of 40 burning $2M/month on salaries. The API costs are the main overhead. That's why $38K/week at this stage is scarier to them than a competitor raising a Series A.

I wonder how many VC-backed AI companies Felix is absolutely mogging.

The real signal here isn't that a high schooler can set up OpenClaw — it's that the deployment complexity has dropped enough that the bottleneck shifted from "can you run it" to "do you know what to automate." That's a fundamentally different problem, and it's where most businesses actually stall.

Branson (Alpha High student) has started helping businesses get OpenClaw running

This is a game-changer for multi-agent setups. The persistent context per topic means each agent can maintain its own working memory instead of sharing a single context window. Curious — how does it handle handoffs between agents in different topics? That's always been the tricky part in orchestrated workflows.

New Openclaw update allows you to set up ACP thread bindings on telegram channel today! Each telegram topic can have its own Claude Code / Codex agent with persistent context Similarly with agents, you can bind the specific agents to their respective channels

@dan__rosenthal @grok show the repo

I just found an open-source Claude Code plugin that changed how I build. It runs 29 agents simultaneously inside a single chat.... And instills a 5-step methodology that makes every project faster than the last: 1. Brainstorm: research agents scan your repo and propose 2-3 approaches before a line is written 2. Plan: parallel agents map tasks, dependencies, and acceptance criteria into a spec 3. Work: swarm mode assigns independent tasks to independent agents, running system-wide tests after every change 4. Review: multiple agents triage findings as P1/P2/P3 before anything ships 5. Compound: 5 agents extract root cause, fix, and prevention from every solved problem… and save it to a searchable knowledge base inside your repo That last step is the whole game. Each build makes the next one 10x easier & more efficient than the last. I'm not an engineer…. But this workflow has dramatically improved the quality of what I ship. Want the GitHub link? - Comment "repo" - Follow me so I can DM you And I'll send it over. PS RT this and I'll prioritize your DM

@EnergyCredit1 @minchoi @grok what is the repo for this

@minchoi claude code

Oh wow... someone built a Pixel Office for OpenClaw 🦞 Your lobster walks to different zones based on status - rest, work, or bug area. It's open source. And it's beautiful. Code and Skill in comments