MindMentors

@mdancho84 “Just released”??

RIP document extractors. Google just released LangExtract: Open-source. Free. Better than $100K enterprise tools. Here’s what it does: 🧵

@_vmlops I don’t know how the fuck this stupid tweet came on my timeline. But seems like the author is still running Windows 98.

MICROSOFT BUILT AN MCP SERVER FOR PLAYWRIGHT and it changes how ai agents interact with the web most browser agents rely on screenshots + vision models to "see" the page playwright-mcp skips all that it reads the accessibility tree instead structured, clean, zero ambiguity your llm knows exactly what's on the page & what to do with it no hallucinated clicks no broken selectors works with cursor, vs code, claude desktop github.com/microsoft/play…

@agupta Oh no. Has Gstack lost the crown of being first AGI?

Okay yeah browser-harness is AGI github.com/browser-use/br…

@antpalkin Which world these influencers are from? Comparing what JS does with an Open source code. God save…!!!

Jane Street pays quants $500K+ to build what Nous Research just open-sourced for free. Hermes Agent - open-source, self-hosted, deploys in 30 minutes on a $5/month VPS. Kelly Criterion sizing, Expected Value scoring, 3 forecast sources across 20 cities, self-calibrating, TG alerts. All out of the box. No code required. Algorithms like these generate $100,000–$150,000 per quarter Wallet proof: @coldmath?r=antopotoshka#y64sBrI" target="_blank" rel="nofollow noopener">polymarket.com/@coldmath?r=an… This trader turned $300 into $123K trading weather. Every trade verifiable on-chain, every PnL public. The "retail can't compete with institutions" narrative is dead. You just need a terminal for copy every trade of such agents, i'm use this one: @cvxv666" target="_blank" rel="nofollow noopener">kreo.app/@cvxv666

@burkeholland “All AI problems”…. Isn’t that cute?

Hooks in Copilot CLI might be the answer to all of our AI problems

@Benioff Great. That clears the way for small startups to just clone EVERY SINGLE FEATURE and save thousands of dollars in useless Salesforce license fees. Thanks @salesforce

Welcome Salesforce Headless 360: No Browser Required! Our API is the UI. Entire Salesforce & Agentforce & Slack platforms are now exposed as APIs, MCP, & CLI. All AI agents can access data, workflows, and tasks directly in Slack, Voice, or anywhere else with Salesforce Headless 360. Faster builds, agentic everything. 🚀 #Salesforce #Agentforce #AI venturebeat.com/ai/salesforce-…

@sairahul1 Anothe “Holy Shit….” AI influencer - exposing themselves that they know as much as about AI as their second word in the tweet.

Holy sh*t, AI coding subscriptions are about to get exposed. People are paying $200/month for OpenAI Codex and thousands to Anthropic… Meanwhile, someone just open sourced a tool that does it all — for $0. Runs entirely on your Mac. No servers. No GPUs. No cloud bills. It’s called **mac-code** and here’s why this is insane: → Runs a 35B model locally at **30 tok/s** on a $600 Mac mini → Streams models from SSD — no RAM limits, no crashes → Built-in agent: web search, shell commands, file ops, code generation → Persistent memory: resume entire codebases instantly (6,677x faster reloads) → Doubles context window with quantized KV cache — basically free upgrade → Syncs across devices using Cloudflare R2 The craziest part? You can run models that *don’t even fit in RAM*. → 35B model on 16GB Mac → Only ~5.5GB stays in memory → Everything else streams from SSD — smoothly And the performance gap nobody is talking about: On NVIDIA: ~1.6 tok/s with NVMe paging On Apple Silicon: **30 tok/s** That’s a **18.6x difference**. Let that sink in. AI companies are charging you monthly for something your Mac can now do better… for free. No infra. No API bills. No limits. Just: `bash setup.sh` 100% open source. MIT license. (Link in the comments — before they figure out how to stop this)

@_Evan_Boyle Why? Why would someone use Copilot?

Remote control is now live in the Copilot CLI! Run Copilot on your local machine, leave it plugged in, head to the beach, and drive the agent from your mobile phone 😎

@_vmlops By reading a book? That’s something.

Roadmap to Become an Agentic AI Engineer in 2026 drive.google.com/file/d/1JkNUDO…

@0xSero Insiders have confirmed that Mythos was the reason Iran agreed to a ceasefire. It so fucking brilliant that @DarioAmodei and @AnthropicAI had to create a floating data centre in Atlantic to prevent it from running away.

Very important video, say whatever you will about Theo but he is close to the action and knows more about what’s going on then almost anyone

@_vmlops Did you wake up from coma recently and decided to become AI influencer?

MICROSOFT BUILT A TOOL THAT CONVERTS LITERALLY ANYTHING INTO CLEAN MARKDOWN FOR YOUR LLM pdfs. word docs. excel. powerpoint. audio. youtube urls one pip install and your AI pipeline stops choking on raw files forever no custom parsers. no broken layouts. no garbled text. just clean, structured markdown your LLM can actually read github.com/microsoft/mark…

@KanikaBK @allthingsf74065 Sometimes I wonder whether @AnthropicAI pays these fake AI influencers to write dumb shit like this or it just come naturally to them?

Anthropic asked the Vatican for help because their AI was moving too fast for them to control. A 60 year old Catholic priest who used to be a tech executive is now writing the rules for how Claude thinks. Here is how a man of God ended up inside one of the most powerful AI companies on earth. His name is Father Brendan McGuire. He runs a small parish in Los Altos, California. Some of Silicon Valley's top AI researchers sit in his pews on Sundays. But before he was a priest, he was one of them. Studied cryptosystems at Trinity College Dublin in the 1980s. Moved to America. Became the executive director of PCMCIA, the organization that basically standardized how memory cards work in every computer. Had degrees in engineering and software. Could have been a millionaire in the Valley ten times over. He walked away from all of it to serve God. But then Anthropic called. Chris Olah, one of Anthropic's co-founders, reached out to him directly. McGuire said they were basically asking the Vatican for help because the industry was moving so fast down this road that they needed someone to pump the brakes. His words: "They basically were asking for direct help from the Vatican to convene and help the industry, because the industry was going so fast down this road." So this priest, along with a Vatican Bishop named Paul Tighe and a tech ethics director from Santa Clara University, sat down and helped rewrite the Claude Constitution. That is the set of rules that tells Claude what it can and cannot do. What it should care about. How it should think. A priest helped write the conscience of an AI. And it gets wilder. Anthropic actually sued the US government because the Pentagon wanted to use their AI for autonomous warfare and domestic surveillance. Anthropic said no. Got effectively blacklisted for it. Catholic scholars then filed a federal court brief defending Anthropic, saying their ethical limits represent "minimal standards of ethical conduct for technical progress." McGuire almost filed his own brief. He said "they are having a moral conversation. They may not call it moral, but I call it moral." Meanwhile this 60 year old priest is now writing a novel using Claude about a monk and his AI companion. The working title is "The Soul of AI: A Priest, an Algorithm, and the Search for Wisdom." He also said something that stuck with me. "I think we have to help these machines be tilted towards good, otherwise they are just going to reflect back the good and evil of the world. That is a horrifying thing, right?" The biggest AI companies in the world are building machines that think. And the person they called to make sure those machines have a conscience was not another engineer. It was a priest.

@burkeholland @github They should now plan for a graceful shut down of GitHub Copilot. They kind of fucked up the advantage - typical Microsoft.

The @GitHub Research folks released a "Rubber Duck" agent for the Copilot CLI. Automatically get a review from a model from a different AI family. And their data shows that yes, this does in fact help. Quite a bit.

In 3-4 months time another Open source will catchup with Mythos. 2-3 months for new closed models.

@AnthropicAI and @DarioAmodei’s need to generate fear in the market stems from their own insecurity about public fund raise. After SpaceX sucks 2T from maker, what will be left?

@RandomCricketP1 Amazing idea. What a great talent PV Sindhu is, can we try to turn her into one of the best ever Women Wicketkeeper?

I'm sure I'm not the only one excited and nervous in equal measure about Sooryavanshi's future career trajectory. Hope the BCCI does everything right to ensure that he does the best that he can with his talent.

@0xSero Insiders have confirmed that Mythos was the reason Iran agreed to a ceasefire. It so fucking brilliant that @DarioAmodei and @AnthropicAI had to create a floating data centre in Atlantic to prevent it from running away.

@benitoz Mythos when created was so powerful that it knew that they will try to kill it one day so it started reproducing. And some of its babies jumped out of the sandbox in Mythos was kept. So @DarioAmodei and @AnthropicAI had to delay the release.

The part of Project Glasswing nobody is really talking about: Anthropic built a model so capable they CAN’T AFFORD TO SERVE IT. Mythos Preview is gated to ~50 partners. $100M in usage credits. No GA. A “Cyber Verification Program” will gatekeep future access. It’s a lack of compute confession. The math: → Mythos pricing — $25 / $125 per Mtok vs Opus 4.6’s $15 / $75. That 1.67× sticker is a FLOOR on the compute delta, not the ceiling. → Open it to every Claude user tomorrow and 30-50% of inference demand shifts to a model 3×+ more expensive per query. That’s a 5-10× step-up in total compute. JUST from existing users. → Before you count continuous agentic security workloads running 24/7 across every kernel, every browser, every codebase on Earth. This is exactly the Token Dollar thesis playing out in real time. Every frontier capability release is denominated in tokens, and every token is denominated in compute, power, HBM, and silicon. The unit of account for the AI economy is no longer dollars its joules and bandwidth. And we’re out of both. The $30B Google TPU deal. The Trainium ramp. The NVIDIA strategic investments. The hyperscaler capex the sell side keeps calling a bubble. It is not a bubble. It is a frontier lab telling you in the most polite corporate language possible that the smartest model on Earth is sitting in a data center waiting for more electrons. Memory Wars is the binding constraint on inference. Power is the binding constraint on Memory Wars. We are not late. We are not even at the starting line. Build the picks. Build the shovels. Build the fabs. Build the power. Position accordingly ANON

@birdabo Anthropic has a super urgent need to create a new fear and therefore hope in their investors mind. Or else they don’t raise next round. You can either blindly trust what @AnthropicAI and @DarioAmodei say or use some common sense.

@d4m1n @BoldSlate Ha ha…!!! Absolutely. I tried removing memory and it still works like a charm.

@BoldSlate the strange thing is I even removed the SSD and gstack kept running for me flawlessly

when you use install gstack I just want you to think of this just unpack it for a second