Carnage
@0xCarnagee • 2,334 subscribers
AI content creator | Tools & automation | @zscdao
Videos
SpaceXAI engineer (ex-Cursor): "I've got 15-25 GrokBot agents running right now. they cover most of what used to be my week there's a Chief of Staff sitting on top of them. it knows every other bot I have and routes the work itself" in a 50-minute podcast, a SpaceXAI engineer walks through building a team of agents that keeps working while you sleep worth more than a $500 course on agentic engineering watch today, then read how to build your own Grok agent team from scratch in the article below
Carnage448,849 次观看 • 1 天前
SpaceXAI senior engineer just released a 1-hour workshop on running Grok 4.6 in Cursor and handing the rest to a GrokBot team - from scratch: • 01:14 - cutting token spend • 04:27 - harness engineering • 09:38 - model amnesia • 30:12 - how to build agents right • 44:08 - demo time: the whole stack running live This 1-hour watch will change how you think about delegating work to agents Skip Netflix tonight, give it an hour, then read the article below on automating your whole routine
Carnage175,514 次观看 • 5 天前
Prime Intellect engineer: "everyone's bragging about a million-token context. here's what they don't tell you. at 256k tokens GPT-5.5 scores 80% on retrieval. push it to a million and it drops to 36%. the model accepts the context, it just can't reason across it. people call it context rot." in a 20-minute talk he explains why bigger context windows won't save your agents. continual learning + training on your own traces + real environments - that's the fix. Watch the talk, then save!
Carnage748,728 次观看 • 1 个月前
ex-Apple engineer gave Grok 4.6 - 30 minutes to redesign a live app, then showed the GrokBot run that merged its own PRs: "GrokBot runs 20-30 agents in parallel while Hermes and OpenClaw run one. Same weights, 20x the throughput - and the agents keep improving from their own runs." Grok 4.6 runs 5x cheaper on input and 8x on output - $2 and $6 per million. his verdict: this is an easy Opus replacement worth more than any $500 course on agentic engineering watch today, then read the article below on building a self-running agent stack
Carnage82,071 次观看 • 6 天前
SpaceXAI engineers just released a 1-hour workshop on running Grok 4.6 in Cursor the way their own team does it - from scratch: • 02:18 · how the context engine works • 11:42 · agents and plan mode • 24:05 · context across big repos • 37:30 · picking the right model • 48:53 · multi-repo without losing it This 1-hour workshop is worth more than 100 articles on Cursor Skip Netflix tonight, give it an hour, then read the article below on wiring Cursor into a GrokBot stack
Carnage44,976 次观看 • 3 天前
Andrew Ng just dropped a free course on Claude Code from scratch, taught with the Anthropic team: 00:00 - why Claude Code is so agentic 04:00 - shockingly simple architecture 12:00 - point it at any codebase this short watch will replace 10 paid coding agent courses. Andrew Ng calls it his personal favorite coding assistant right now. Watch it today, then read how to engineer your own agent loops in the article below
Carnage711,239 次观看 • 1 个月前
Anthropic's CEO Dario Amodei: "we saw 3x revenue growth in a single quarter. annualized that's 80x. we didn't plan for 80x. it would not have been rational" "my son asked to use my Cowork account. I said absolutely not. I need my tokens" on Mythos: "companies we gave it to said – this is a super weapon. you should have to own a gun license to use it. please don't release this" it found 271 new vulnerabilities in Firefox. thousands more in private companies that haven't been disclosed yet "we have suffered enormously commercially from not releasing Mythos. it would accelerate everything. but we won't" bookmark & watch ↓
Carnage942,932 次观看 • 2 个月前
Anthropic's CEO Dario Amodei: "some of the early companies we gave Mythos to said – this is a super weapon. please don't release this" on leaving OpenAI: "when you feel that you can't trust someone, when you feel that their values are not what they say they are – that makes it very hard to continue" Boris Cherny, Head of Claude Code: "90% of code at Anthropic is written by Claude. a lot of Claude Code's code is written by Claude Code" Dario puts the odds of civilizational collapse from AI at 10–25% 47 minutes. free. Dario, Daniela, Boris – the deepest look inside Anthropic ever filmed bookmark & watch ↓
Carnage740,646 次观看 • 2 个月前
Anthropic's Head of Claude Code: "I don't have a to-do list anymore. Claude just builds everything" he runs a tree of thousands of agents. agents prompting agents prompting agents one engineer set up a routine that auto-fixes any bug report not responded to in 5 hours. merges the easy ones himself Boris shipped a feature with an edge case. before he could fix it - another engineer's routine already had a PR up "Claude tells me this all the time. someone else has already fixed it. there's always another Claude working on it" 17 minutes. free. the most honest look at how the Claude Code team actually works bookmark & watch ↓
Carnage374,140 次观看 • 2 个月前
SpaceXAI engineer (ex. Cursor): "i've got 10-20 GrokBot agents running right now. they handle about 90% of the routine there's a Chief of staff agent on top - it knows every other bot I have and runs the whole thing" in a 35-workshop, a SpaceXAI engineer walks through building a team of GrokBot agents that works around the clock Worth more than a $500 course on agentic engineering Watch today, then read how to build your own Grok agent team from scratch in the article below
Carnage33,121 次观看 • 10 天前
Elon Musk & xAI core team: "Grok is a strong model. The interesting part is what comes after it GrokBot is where that goes - agents that improve from their own runs instead of waiting on the next model" In a 20-minute talk, Elon and the SpaceXAI engineers lay out where AI engineering goes next worth more than 2 hours of Stanford lectures on agentic engineering watch today, then read the article below on building a self-learning agent system with GrokBot
Carnage20,217 次观看 • 7 天前
xAI co-founder and chief engineer: "you have your Grok bot agent attempt many solutions, you collect the rewards then you use reinforcement learning to update the weights. that loop is the whole thing" he built Colossus, the cluster everyone else quoted at over a year, by refusing to hire subcontractors of subcontractors on the Cursor acquisition: the data matters, but the real value is the RL environments - hand the agent a bug, then check whether it fixed it and whether it fixed it the right way Watch the talk, then read the article below.
Carnage27,541 次观看 • 11 天前
SpaceXAI engineers (ex-Cursor): "Right now I've got 15-30 GrokBot agents running. They close 100% of my code bugs while I'm asleep every time feedback comes in on GrokBot, my bots pick it up and go fix it themselves" In a 28-minute, SpaceXAI engineers walk through agentic engineering Worth more than 100 articles on setting up GrokBot, straight from the engineers who built it in the article below Watch today!!!
Carnage23,035 次观看 • 9 天前
Anthropic's Head of Claude Code Boris Cherny: "Claude Code went from zero to $400 million in revenue in just a few months. it started as a hackathon project. it wasn't even meant to be a product" it still only uses the public Anthropic API. same models everyone else uses. "dog fooding successful" "over time I would actually expect cost to increase. the only thing we care about is capability. if we have to compromise capability, we won't do it" sub agents came from a Reddit post. someone built an army of agents PM, designer, frontend, backend. an engineer saw it and built it during a hackathon. it took off like wildfire internally 30 minutes. free. at Bessemer Venture Partners bookmark & watch ↓
Carnage162,430 次观看 • 2 个月前
Karpathy just proved the structure beats the model - mapped the whole path from loops to graphs into a 15-page PDF: he pointed one agent at code he'd already hand-optimized. 700 experiments in two days, then he said the single loop was never the target here's the full system: step 1 → write the scorer before anything else. one number, automatic, under 10 minutes. no scorer, no loop step 2 → split the files. one the agent edits, one holding the scorer it can never reach step 3 → fix the wall clock per attempt. comparability comes from the budget, not the attempts step 4 → write the never-stop line verbatim or your agent becomes a slower human step 5 → add a second agent only after the first ships real commits. one git worktree each or they overwrite each other step 6 → at thirty agents, isolation costs more than it saves. that's when they need one shared DAG instead of thirty private histories 2.02 hours to 1.80 hours on code its author had already tuned. 90,900 stars in four months. read this PDF and paste it into your Claude, you won't regret it bookmark, then read the article on building graphs from scratch ↓
Carnage53,872 次观看 • 1 个月前
Andrej Karpathy to a room of AI agent builders: "a demo is easy. a product takes a decade, self-driving was easy to demo in 2016, easy to imagine, and it still took ten years to ship, agents are the same" in a short talk he tells builders the thing the hype cycle won't easy to imagine + easy to demo + brutally hard to ship - that's the trap Watch the talk, then read the article below
Carnage26,872 次观看 • 1 个月前
没有更多内容可加载