Amjad Masad
@amasad ·
Massively underreported: an official Claude Code extension was compromised and installed backdoors on developer machines.
Could it be behind the Binance hack?
Back
AI posts
AI posts
People and companies shaping AI. Filter your watched accounts, catch up on recent posts, or find the most liked and viewed.
Compiled by the Toolstash team · Updated
209 watched accounts · 2988 matching posts · Showing up to 3,000 recent stored posts. Counts are snapshots, not live X totals. This page refreshes every minute while visible. Last collected .
Andrew Curran@AndrewCurran_ ·
Looks like Anthropic are cracking down on Claude Code videogame mashups. I've seen some amazing ones posted in group chats. The way this has progressed since September, it will soon be as easy as mixing together two colors of paint.
Mike Taylor@hammer_mt ·
Even with a model as smart as Opus 5.5 there's still unreasonable amounts of alpha in starting a new session and copying the link to the old session to ask what could have been done better. It identified that claude was re-rendering everything for every single change I requested (huge waste of tokens and time!) when making this video. Shows that there's still alpha in creating your own skills / CLIs to stop claude doing wasteful stuff. Don't listen to the model maximalists who tell you skills aren't useful or prompt engineering is dead.
Daniel San@dani_avila7 ·
Yes! When you open Rex, two programs start:
- The app you see - server running in the background
The server keeps your sessions alive even when you close the app
Elon Musk@elonmusk ·
The Democrats intend to cement one-party rule forever and make “democracy” meaningless, as they have done in California
- Elon Musk
@elonmusk ·
Four years ago …
And that is exactly what happened
Alex Lieberman@businessbarista ·
I'm interviewing one of the world's experts on cybersecurity today (at 12pm ET).
1) The history of cyber 2) How cyber worked traditionally 3) How AI changed cyber 4) How post-AI cyber works 5) What every enterprise needs to do right now
Most professionals are woefully underinformed about cyber & need to get smart fast.
RSVP here: https://luma.com/tenexlabs-ztrt
Tibo@tibo_maker ·
incredible to see people switching from a competitor to Bazzly and being happy with the results 🤯
those people know the importance of the Reddit game and having them satisfied is GOLD
jack friks@jackfriks ·
wake me up when i can run opus 5.5 on my macbook locally thank you and see you soon (2027)
elvis@omarsar0 ·
The era of hardcore fundamental research is back!
This is key to improving AI efficiency.
The brain runs on roughly 20 watts, so it makes sense to study how it computes.
@BioComputingCo has been doing that with living neurons for four years. Their first product is mischief 1, a video model at least 5x faster than its base model.
I'm curious to see where this research goes after video.
Ethan Mollick@emollick ·
This is right, but I'd add that the distance between what someone who is working at the frontier with AI uses AI for and what AI can actually do today is growing even wider! We don't really understand the full capability of today's systems when properly harnessed & applied.
Peter Gostev@petergostev ·
What's the bull case for bitcoin now? That hackers will get bored of trying to get at the hundreds of billions of wealth sitting around there for the taking?
- Tibo
@tibo_maker ·
weekend side-project:
1. check your unused domains 2. connect them to Outrank Marketplace (free) 3. start earning 💸
fully automatic then you collect the payouts already $26k+ earned
start here: http://outrank.so/sell
(disclaimer: we will validate each domain, some may be rejected like if DR is lower than 30)
- Peter Gostev
@petergostev ·
UI solved by before the end of the year?
nicksaraev@nicksaraev ·
I feel for people like this. They've spent their whole lives preparing for a world that will never exist.
- jack friks
@jackfriks ·
good morning, i love you @cloudflare btw
@levelsio@levelsio ·
What you see happening in France is essentially what will happen throughout the whole of Western and Southern Europe at some point
Europe regulated away all new and existing businesses so they with the money and the talent moved elsewhere (like US and Asia)
When money flees a continent, the businesses, then the people and the government run out of money
Meanwhile cost of living keeps growing → people get poorer and → people need more government welfare → meaning the government has to borrow more money to pay for that → borrowing for worse and worse rates → which pushes down the Euro → meaning less spending power for its people → making Europeans even poorer
Santiago@svpino ·
This is a great model for exploring ideas quickly and cheaply.
It's an image-to-video generation model. 5x faster and 80% lower inference cost than its base open-weight model.
• Input: a starting image and a prompt describing what you want • Output: a 720p video • Advantages: Fast generation, low cost. Easy to iterate. • Disadvantages: Output doesn't have production quality.
Marc Lou@marclou ·
TrustMRR a year ago:
- a vibe-coded leaderboard
TrustMRR now:
- 4,000,000 visitors - 208 startups acquired - $427,610 in revenue (90% margin) - several multi-million acquisition offers rejected
Happy 1st birthday, https://TrustMRR.com 🎂
Peter Steinberger 🦞@steipete ·
Another reason why you wanna use your own harness and not one where companies change what you can or can’t do at will.
Vox@Voxyz_ai ·
19 AI coding rules, all from 14 of my posts that 𝗲𝗮𝗰𝗵 𝗽𝗮𝘀𝘀𝗲𝗱 𝟮𝟬𝟬𝗞 𝘃𝗶𝗲𝘄𝘀. Together those posts have 8.7M+ views and 94K bookmarks.
Last night I went back through six months of posts, broke them down rule by rule, rewrote each one, and took several drafts to get to the list below.
Kickoff 1. When you hand it a task, give it a goal: how far to take it, how to verify each step, and what counts as done. 2. When unsure, ask first. Don't silently pick one interpretation and run with it. Touch only what's required, and don't fix up neighboring code on the way. 3. For any task over 5 steps or 30 minutes, have it build a progress dashboard before it starts and update it after every step: progress, what's stuck, questions waiting on you, and what it'll do by default if you haven't answered. In Claude Code, make it an artifact: the page refreshes in place, and Ctrl+] opens it anytime. In other tools, a local HTML page works.
In AGENTS.md / CLAUDE.md 4. Delete lines like "think carefully" and "take a deep breath" that push it to think more. Opus 5.5 already thinks before every reply and decides how much. Delete them and 𝗿𝗲𝗽𝗹𝗶𝗲𝘀 𝗰𝗼𝗺𝗲 𝗯𝗮𝗰𝗸 𝗳𝗮𝘀𝘁𝗲𝗿, with no drop in quality. 5. If code is only used in one place, don't pull it out into an abstraction, adapter or config layer. Add one when a second place actually needs it. 6. Don't design for rare or future cases nobody asked about. Delete the old code you replace, unless compatibility is an explicit requirement. 7. Before adding a dependency, changing a public API or data format, or deleting user data, stop and ask me. 8. When you switch models, have the new one review your AGENTS.md and skills against the official guide: overly broad triggers, docs it has to read for every task, redundant checks, conflicting instructions, and rules that make it stop and check with you before the work is done. Recommendations first, no edits yet.
UI 9. Give it a DESIGN.md first that spells out colors, type sizes, spacing, corner radius, shadows and component rules, and have it read the file before every UI build. Don't want to write one? Take a ready-made DESIGN.md from a product you like and drop it in your project root. 10. Codex has image generation built in. If you think the UI it made is ugly, tell it: "Use imagegen to reimagine this page, then implement it." 11. Pick one page and have it redo it in the style of a product you like. Pin down the style first: a restrained palette, a clear type hierarchy, plenty of whitespace. Visuals only, no new features, and have it walk you through every change. 12. Cut taglines, helper text and secondary labels. If a screen needs them to make sense, fix the layout.
When you're stuck 13. When fixing a bug, have it find the root cause and explain the fix before it touches the code. 14. If Opus 5.5 is your main model, run /advisor fable to put Fable on the sidelines as an advisor. Before committing to a plan, when the same error shows up a second time, and before saying it's done, Opus checks with it first. Opus still writes the code.
Before you ship 15. Have Opus 5.5 send out 5 subagents to check the 20 things worth checking (design, mobile, plus the loading, empty and error states), then fix them all in one pass. 16. Then send 4 subagents after the 20 things that slow a site down. Run Lighthouse in mobile mode three times before and after the fixes and compare the medians. Don't cut features to gain speed. 17. Last, have it play the bad guy: with two regular user accounts and one admin account, actually run all 14 attacks, keep the evidence, then fix what it finds.
Saving usage 18. Start effort at medium: low for mechanical edits, medium for everyday work, high when medium gets stuck, and save xhigh and max for work where you've measured a gain. 19. Subagents that only search, read logs or run tests need their own model, or they all run on your main model. Unused MCP servers and a CLAUDE.md well over 200 lines 𝗲𝗮𝘁 𝘂𝘀𝗮𝗴𝗲 𝗲𝘃𝗲𝗿𝘆 𝘁𝘂𝗿𝗻. Turn those MCPs off and cut CLAUDE.md down.
Each rule's original post is linked in the first reply. The full checklists and prompts are in those posts.
Maziyar PANAHI@MaziyarPanahi ·
i clicked once on the liver in a public abdominal CT and it popped out in 3D 🔥
then the organs around it, then everything: 74 structures in 13 seconds with the open NV-Segment-CT model on one GPU
surgeons, would you plan a liver resection with this, or stay on the slices?
- Ethan Mollick
@emollick ·
One thing we don’t have a good measure of is the gap between the capability of the best internal models at the labs and the best released model to the public. I suspect it has been growing quite a bit over the course of the last year, especially after the temporary Fable ban.
Josu San Martin@josusanmartin ·
The Binance hack was equivalent to 10~15% of North Korea's estimated GDP.
Mia@MiaAI_lab ·
Absolutely insane decode numbers on 2x RTX PRO 6000s with GLM 5.3 Flash on TensorFold. Haven't even optimized it really... faster than API I think.
This is with power capped at 400w per card.
- Santiago
@svpino ·
People are celebrating how much time they're saving by letting agents review their code.
But be honest: how much time are you spending fixing bugs afterward?
Saving two hours on code review means nothing if you then spend four hours fixing problems.
If you really want to know whether your automated code reviews are saving you time, you gotta measure the whole thing:
1. Track how much time you spend reviewing code, including the agent's feedback.
2. Track how much time you spend fixing bugs that make it past review.
3. Track how much time you spend investigating incidents and dealing with affected customers.
Compare the total time with what it took before introducing agentic reviews. Ideally, measure it per PR or per change so you're comparing similar workloads.
At the end of the day, nobody cares how much time you are saving on a particular task.
The ultimate goal is to save more time overall without deploying crap to production.
- Josu San Martin
@josusanmartin ·
“The outflows ceased when we paused withdrawal service earlier.”
No, they didn't. The hackers continued draining funds for hours after withdrawals were paused.
Making such an easily verifiable false statement about a security incident doesn't exactly inspire confidence.
- Amjad Masad
@amasad ·
If it’s made of code, we can run it!
Paulos Yibelo@PaulosYibelo ·
I predict people will run back to popular mature software once the hacks really start (and no, they haven't started. AT ALL). Yeah you can use a sandbox this guy vibe coded but you'll learn a painful & expensive lesson & run to Firecracker/gVisor, same for the browser, the framework, the infra, the kernel, the web server, or anything. If you think KVM or Chrome having a couple rce means you should go build your own, or use this other new thing, you have no idea how bad it's going to get for you. As weird as it may seem, I do believe software will actually centralize & consolidate into few winner products due to security reasons. If you want to win in this coming era, BUILD SECURE SOFTWARE. That will be a moat at a time anyone can build anything - but It's not as easy as running Mythos on your code. Hackers are going to hack. There isn't a coverage checklist you go through and you'll be safe (although that's the dominant bs cybersecurity "expert" narrative). You will never be safe, you are only safer than the next guy. Assume breach, hire real hackers & be paranoid beyond reason. Building secure software is not easy but the bar is too low to compete - esp for technical leaders. If your security tests are people you employ who are mediocre, you are going to get hacked, no matter your threatmodel. Realistically most can't hire the few best hackers out there but if you want to win by remaining out of the news when your competitor product gets inevitably gets hacked, spend the money and get even 1 hour of their time. This will make sense in due time. b660484e4727dca6c5c08293bc29e832df2b151253adeffd34163d3a5217de7e
- Mike Taylor
@hammer_mt ·
I have a skill for writing up eval experiments as artifacts and just added these principles to it. It was already pretty well optimized but this still really helped. Here's the before and after:
