- Published on
科技推文精选 - 2026-03-29
- Authors
- Name
- geeknotes
2026年3月29日科技每日简报
Today's top tech conversations are led by @chhddavid, whose post about '$270k. that's what it used to ...' garnered the highest engagement. Key themes trending across the top stories include claude, anthropic, coding, model, about. The community is actively discussing recent developments in AI, engineering practices, and startup strategies.
1. chhddavid (Group Score: 139.3 | Individual: 37.3)
Cluster: 4 tweets | Engagement: 53 (Avg: 22) | Type: Tech
$270k. that's what it used to cost to make a website look like real expert humans made it.
thats over.
I've been vibe coding websites and mobile apps for over a year and I honestly believe there's no difference between "indie hacker" and "VC-backed SaaS with 35 employees" anymore...
but this is different. type out your thoughts and ai... adapts. design, code, monetization, launch, email marketing - you're not doing anything anymore, just approving/disapproving AI's decisions
the skill now is SHIPPING. knowing what sticks and what doesn't, then doubling down on what actually works
an old friend of mine has 0 technical ability. literally none. created his own calorie tracking app two weeks ago and it's better than half the agency work I've seen the last 6 months. took him <18 hours...
and he didn't code/design/write a single thing. he just took an already-validated idea and made it his own...\n\nQT @chddaniel: Today is the end of vibe coding...
I just watched my Mac build a company in 193 seconds.
This is absurd. https://t.co/2U0lvi4InA
See 3 related tweets
- @chddaniel: $63,800/mo.
that's what it used to cost to make an full platform look like real expert humans made ...
- @chddaniel: $63,800/mo.
that's what it used to cost to make an full platform look like real expert humans made ...
- @chhddavid: $270k. that's what it used to cost to make a website look like real expert humans made it.
thats ov...
2. adamlyttleapps (Group Score: 104.3 | Individual: 40.1)
Cluster: 3 tweets | Engagement: 608 (Avg: 151) | Type: Tech
Screw it, I made it open source..
This is Notchy -_-
He stops you getting distracted when using Claude code by replacing your Macbooks notch with a terminal
He lets you know when claude needs your attention And plays a sound when tasks are complete
Best of all: he stops your macbook going to sleep while claude is working
I built this for me, maybe you will find it useful too?
As a swift developer Notchy has some custom functionality I built:
- When a new XCode project is open he launches a new tab
- If claude.md is detected he launches straight into claude code
- Command + S saves a quick snapshot of code and I can restore from that checkpoint any time
Enjoy :)
https://t.co/bImzV9tWJx\n\nQT @adamlyttleapps: I kept getting distracting while vibe coding… so I made a notch for Claude Code
It updates the status, pings you when you need to answer a question and notifies you when the task is done
When it detects claude is working it also prevents my macbook from going to sleep
I can walk away from my macbook. Or watching a youtube video. And I'll get an alert when it's done.
See 2 related tweets
- @adamlyttleapps: New business model:
Release free open source tools
Put together a short launch video
Profit from ...
- @adamlyttleapps: Silly projects are back
And I'm loving it https://t.co/v8SsRfCS4W\n\nQT @adamlyttleapps: Screw it, ...
3. AndrewCurran_ (Group Score: 101.7 | Individual: 38.1)
Cluster: 3 tweets | Engagement: 1608 (Avg: 803) | Type: Tech
Three weeks ago there were rumors that one of the labs had completed its largest ever successful training run, and that the model that emerged from it performed far above both internal expectations and what people assumed the scaling laws would predict. At the time these were only rumors, and no lab was attached to them. But in light of what we now know about Mythos, they look more credible, and the lab was probably Anthropic.
Around the same time there were also rumors that one of the frontier labs had made an architectural breakthrough. If you are in enough group chats, you hear claims like this constantly, and most turn out to be nothing. But if Anthropic found that training above a certain scale, or in a certain way at that scale, produces capabilities that sit far above the prior trendline, then that is an architectural breakthrough.
I think the leaked blog post was real, but still a draft. Mythos and Capybara were both candidate names for the new tier, though Mythos may now have enough mindshare that they end up keeping it. The specific rumor in early March was that the run produced a model roughly twice as performant as expected. That remains unconfirmed. What is confirmed is that Anthropic told Fortune the new model is a 'step change,' a sudden 2x would certainly fit the definition.
We will find out in April how much of this is true. My own view is that the broad shape of this is correct even if some of the numbers are wrong. And if it is substantially accurate, then it also casts OpenAI's recent restructuring in a new light. If very large training runs are about to become essential to staying in the game, then a lot of their recent decisions, like dropping Sora, make even more sense strategically.
For the public, this would mean the best models in the world are about to become much more expensive to serve, and therefore much more expensive to use. That will put pressure on rate limits, pricing, and subscription plans that are already subsidized to some unknown degree. Instead of becoming too cheap to meter, frontier intelligence may be about to become too expensive for most of humanity to afford.
Second-order effects; compute, memory, and energy are about to become much more important than they already are. In the blog they describe the new model as not just an improvement, but having 'dramatically higher scores' than Opus 4.6 in coding and reasoning, and as being 'far ahead' of any other current models. If this is the new reality, then scale is about to become king in a whole new way. It would also mean, as usual, that Jensen wins again.
See 2 related tweets
- @WesRoth: great read\n\nQT @AndrewCurran_: Three weeks ago there were rumors that one of the labs had complete...
- @dejavucoder: big if true https://t.co/mxh4GRpSbl\n\nQT @AndrewCurran_: Three weeks ago there were rumors that one...
4. _xjdr (Group Score: 96.7 | Individual: 38.9)
Cluster: 3 tweets | Engagement: 563 (Avg: 151) | Type: Tech
calling Nicholas Carlini "someone at anthropic" instead of "one of the best security researchers alive" is very funny to me\n\nQT @chiefofautism: someone at ANTHROPIC just showed CLAUDE finding ZERO DAY vulnerabilities in a live conference demo
claude has found zero day in Ghost, 50,000 stars on github, never had a critical security vulnerability in its entire, history...
it found the blind SQL injection in 90 minutes, stole the admin api key, then did the exact, same thing to the linux kernel
See 2 related tweets
- @giffmana: "someone at Anthro"
lol you typo'd
"the most goated security x ai researcher ever"\n\nQT @chiefofa...
- @garrytan: RT @chiefofautism: someone at ANTHROPIC just showed CLAUDE finding ZERO DAY vulnerabilities in a liv...
5. NathanpmYoung (Group Score: 84.5 | Individual: 31.9)
Cluster: 4 tweets | Engagement: 758 (Avg: 127) | Type: Tech
RT @StefanFSchubert: While social media is polarising, evidence suggests AI may nudge people towards the centre.
This holds true of all studied models. Grok is more right-leaning than other models, but also has depolarising effects.
By @jburnmurdoch. https://t.co/Fokx869fVq
See 3 related tweets
- @danshipper: not surprising if you’re paying attention
https://t.co/gxBaac30WG\n\nQT @StefanFSchubert: While soc...
- @garrytan: AI might be a powerful depolarizer in political discourse\n\nQT @StefanFSchubert: While social media...
- @rickasaurus: Some good news\n\nQT @StefanFSchubert: While social media is polarising, evidence suggests AI may nu...
6. badlogicgames (Group Score: 83.4 | Individual: 44.6)
Cluster: 3 tweets | Engagement: 3776 (Avg: 119) | Type: Tech
RT @_chenglou: My dear front-end developers (and anyone who’s interested in the future of interfaces):
I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept): Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow
See 2 related tweets
- @rauchg: I can finally ship the text layout of my dreams https://t.co/feyBNQXxLS\n\nQT @_chenglou: My dear fr...
- @saranormous: Whoaaaaa\n\nQT @_chenglou: My dear front-end developers (and anyone who’s interested in the future o...
7. WesRoth (Group Score: 82.9 | Individual: 33.0)
Cluster: 4 tweets | Engagement: 91 (Avg: 30) | Type: Tech
According to a massive new leak, Anthropic is preparing to unveil a frontier model that transcends its current naming conventions.
Dubbed "Claude Mythos," this new model reportedly sits entirely above the "Opus" tier and represents a fundamental step-change in artificial intelligence capabilities.
The model reportedly demonstrates massive performance gains in highly complex domains, specifically advanced coding, academic reasoning, and cybersecurity.
Because Mythos is so compute-intensive and possesses such advanced capabilities, Anthropic is reportedly not planning a standard public release.
Access will roll out extremely slowly, starting exclusively with select cybersecurity partners to help them prepare for and defend against potential AI-driven exploits.
See 3 related tweets
- @rohanpaul_ai: RT @rohanpaul_ai: Rumor has it Anthropic is getting ready to launch new models, Mythos and Capybara,...
- @minchoi: RT @minchoi: 🚨BREAKING: Fortune says leaked Anthropic docs show "Claude Mythos" is already in testin...
- @rohanpaul_ai: RT @rohanpaul_ai: Fortune just reported about this leak:
Anthropic's new generation of super-stron...
8. SIGKITTEN (Group Score: 80.9 | Individual: 52.1)
Cluster: 2 tweets | Engagement: 418 (Avg: 48) | Type: Tech
RT @badlogicgames: we as software engineers are becoming beholden to a handful of well funded corportations. while they are our "friends" now, that may change due to incentives. i'm very uncomfortable with that.
i believe we need to band together as a community and create a public, free to use repository of real-world (coding) agent sessions/traces. I want small labs, startups, and tinkerers to have access to the same data the big folks currently gobble up from all of us. So we, as a community, can do what e.g. Cursor does below, and take back a little bit of control again.
Who's with me?
See 1 related tweets
- @ivanfioravanti: I'm in Mario!\n\nQT @badlogicgames: we as software engineers are becoming beholden to a handful of w...
9. WesRoth (Group Score: 70.7 | Individual: 35.4)
Cluster: 2 tweets | Engagement: 70 (Avg: 30) | Type: Tech
Google is stepping up to finance a multi-billion dollar data center in Texas specifically built to house Anthropic's AI operations. The sheer scale and energy strategy of this facility are setting a new precedent for the industry.
A massive 1-Gigawatt (1GW+) facility will be powered entirely by on-site natural gas turbines. Rather than relying on the traditional electrical grid, the data center will generate its own dedicated fossil-fuel power.
Nexus Data Centers will operate the facility and lease it to Anthropic. Banks are currently competing to provide up to $5 billion in financing just for phase one of the buildout.
Google already owns roughly a 14% stake in Anthropic, and Anthropic has previously locked in a massive commitment to purchase 1 million of Google’s proprietary AI chips (TPUs) to train its Claude models.\n\nQT @FT: Google nears deal to help finance multibillion-dollar data centre leased to Anthropic https://t.co/aybRZhK7vc
See 1 related tweets
- @rohanpaul_ai: BREAKING. FT reports Google is moving to help fund a $5B+ Texas data centre that Anthropic will leas...
10. garrytan (Group Score: 66.7 | Individual: 34.2)
Cluster: 2 tweets | Engagement: 620 (Avg: 322) | Type: Tech
I have to say this interview changed my life. Hearing how Boris thinks about software spurred me to work much harder on releasing my own way of doing things and on iterating fast on how I build. Hard to believe it has only been a month since this one.\n\nQT @ycombinator: A very special guest on this episode of the Lightcone! @bcherny, the creator of Claude Code, sits down to share the incredible journey of developing one of the most transformative coding tools of the AI era.
00:00 Intro 01:45 The most surprising moment in the rise of Claude Code 02:38 How Boris came up with the idea for Claude Code 05:38 The elegant simplicity of terminals 07:09 The first use cases 09:00 What’s in Boris’ https://t.co/OAtnXdxccP? 11:29 How do you decide the terminal’s verbosity? 15:44 Beginner’s mindset is key as the models improve 18:56 Hyper specialists vs hyper generalists 21:51 The vision for Claude teams 23:48 Subagents 25:12 A world without plan mode? 28:38 Tips for founders to build for the future 30:07 How much life does the terminal still have? 30:57 Advice for dev tool founders 32:11 Claude Code and TypeScript parallels 35:34 Designing for the terminal was hard 37:36 Other advice for builders 40:31 Productivity per engineer 41:36 Why Boris chose to join Anthropic 44:46 How coding will change 46:22 Outro
See 1 related tweets
- @ycombinator: RT @garrytan: I have to say this interview changed my life. Hearing how Boris thinks about software ...
11. testingcatalog (Group Score: 66.6 | Individual: 35.5)
Cluster: 2 tweets | Engagement: 366 (Avg: 223) | Type: Tech
Availability of Seedance 2.0 on CapCut got expanded to a loads of new countries, including EU.
CapCut is already one of the top AI apps and seems like they are only starting.
Some free credits too 👀 https://t.co/baI7OnW5Nd\n\nQT @capcutapp: Today we are expanding Dreamina Seedance 2.0 to more users worldwide within CapCut - including Europe, Canada, Australia, New Zealand, South Korea, SEA, MENA, LATAM and Africa.
Plus, we’ve provided everyone with one free trial of Dreamina Seedance 2.0 across CapCut’s app, desktop and web. Enjoy creating now!
Here’s a quick guide to help you explore different CapCut features that support Dreamina Seedance 2.0: https://t.co/Mgdbr6hykN
RT+Comment in 9hr to get extra 1000 credit in your DM!
See 1 related tweets
- @WesRoth: CapCut is expanding access to its advanced AI video generation model, Dreamina Seedance 2.0, to user...
12. chddaniel (Group Score: 64.9 | Individual: 32.9)
Cluster: 2 tweets | Engagement: 17 (Avg: 36) | Type: Tech
saas in 2024:
> raise $1M > hire a design + marketing agency > get devs to "build in stealth" > launch 17 months late > nobody buys
saas in 2026: > find an already-printing idea > paste it into shipper > launch in 27 mins > get customers the same afternoon\n\nQT @chhddavid: Today, we've ended vibe coding forever...
I just witnessed my Mac build a full company in 182 sec.
This is beyond crazy. https://t.co/xcYIiczDP0
See 1 related tweets
- @chhddavid: SaaS in 2022: > raise 1M > hire a design agency (8k/mo) > get 8 devs "building in stealth...
13. MarioNawfal (Group Score: 63.5 | Individual: 36.6)
Cluster: 2 tweets | Engagement: 663 (Avg: 848) | Type: Tech
🇺🇸 50% of Tesla owners are using Full Self-Driving 90-100% of the time.
Half the fleet is letting the car drive itself… almost always.
That’s real trust. And it’s only going up.
https://t.co/tyr4vezbgl\n\nQT @MarioNawfal: 🇺🇸 Tesla just dropped its boldest vision yet: “Accelerate the world's transition to amazing abundance.”
Scaling tech + energy to make everything cheaper & abundant for all.
Ambitious? Hell yes. Only @Tesla
See 1 related tweets
- @MarioNawfal: RT @MarioNawfal: 🇺🇸 Tesla just dropped its boldest vision yet: “Accelerate the world's transition t...
14. SawyerMerritt (Group Score: 63.4 | Individual: 31.8)
Cluster: 2 tweets | Engagement: 1757 (Avg: 1880) | Type: Tech
Tesla's Optimus team https://t.co/P0aB2OufgK\n\nQT @K0nstantin0s_: A fitting analogy for the pace at which Tesla has been making progress over the years is that of a Formula One car, whose peak acceleration increases in proportion to its speed as a result of enhanced traction generated by aerodynamic downforce.
Our goal is to get Optimus to high-volume production as fast as possible.
If your passion is designing drive systems, electronics, sensors, dynamic harness, cameras, body or hand structures and you want to overcome micron tolerancing challenges, and accelerate the scalability of humanoid robot manufacturing, this is the opportunity for you!
You will learn a ton and make a huge impact!
Apply here: https://t.co/xb16dNyflQ
See 1 related tweets
- @niccruzpatane: New look for Tesla Optimus robot. It literally looks like a human in a robot suit. Wow. https://t.co...
15. rohanpaul_ai (Group Score: 60.9 | Individual: 35.6)
Cluster: 2 tweets | Engagement: 82 (Avg: 83) | Type: Tech
AI data centers did not stop at squeezing Dynamic Random-Access Memory (DRAM) and NAND flash, they have now dragged the CPU market into the mess too.
The AI bottleneck is now also moving from accelerators to the CPUs that keep them fed and synchronized.
A March study found that moving 4 to 8 GPU LLM servers from minimal CPU allocation to CPU-abundant setups cut time-to-first-token by 1.36 to 5.40 times, with some lean configurations timing out entirely.
That is because the CPU now runs the control plane: tokenization, kernel launches, batch scheduling, inter-process messaging, and the storage and networking work that keeps tensor-parallel GPUs busy.
Decode is where this turns nasty: TaxBreak, another March paper, shows autoregressive serving multiplies host overhead token by token, so 10 decode tokens took 188 ms versus 22 ms for prefill in one setup, and faster CPU single-thread performance cut host-bound latency by 11 to 14 %.
The ugliest failures are structural, not just a shortage of cores: the Georgia Tech paper found vLLM’s shared-memory broadcast queue could stretch from about 12 ms to 228 ms under load, making the CPU control path roughly five times longer than the GPU compute step and worsening with tensor parallelism.
That is why the supply squeeze matters. Reuters reported AI-driven server-CPU lead times stretching to six months and prices rising by more than 10 % in some markets, and Intel said in January it was struggling to meet AI-data-center CPU demand.
The interesting twist is that, now Arm may gain ground, because a shortage in x86 server CPUs gives buyers a reason to test alternative chips built specifically for AI-heavy data centers.
Arm’s opportunity is not just that some buyers suddenly like a different chip architecture in theory.
It is that AI data centers now care intensely about practical system traits like memory bandwidth, I/O, power efficiency, and host-side scheduling, so a CPU that handles those jobs well can win even if the buyer was historically committed to x86.
Arm’s new AGI design is pitched around 12 DDR5 channels, more than 800 GB/s of bandwidth, 96 PCIe Gen6 lanes, and CXL 3.0, though its rack-scale claims still need independent proof.
The deeper shift is that AI infrastructure is becoming a coordination problem disguised as a compute problem, and coordination runs on CPUs.
See 1 related tweets
- @rohanpaul_ai: RT @rohanpaul_ai: AI data centers just pushed the hardware crunch beyond DRAM and NAND and into the ...
16. seraleev (Group Score: 60.2 | Individual: 36.2)
Cluster: 2 tweets | Engagement: 141 (Avg: 51) | Type: Tech
These screenshots helped Adam hit $70k/month. Save this formula:
> strong contrast > verb-first headlines (action-driven) > large, readable typography
Save this, you’ll need it later\n\nQT @adamlyttleapps: I'm creating a Claude Skill that makes ASO optimized screenshots for apps
This is my exact workflow for building an app portfolio doing $70k/month
The skill assesses my code. Works out the primary beneifts. Creates the hero headlines.
Then directs you what screenshots from your app it needs. You provide the screenshots from the simulator and it pairs up the correct heading with the right screenshot.
Then:
It puts it altogether. Adds pizazz with Nano Banana
And hey presto
No more photoshop
Should I release this?
See 1 related tweets
- @adamlyttleapps: I'm creating Claude Skill that does it all in 15 mins... https://t.co/1DyDNykhBW\n\nQT @seraleev: Th...
17. jerryjliu0 (Group Score: 60.0 | Individual: 34.9)
Cluster: 2 tweets | Engagement: 119 (Avg: 88) | Type: Tech
Last week we launched LiteParse - a free and fast document parser that provides more accurate AI-ready text than other free/fast parser libraries.
It’s a great tool you can plug into assistant agents like Claude Code/OpenClaw and get good results, especially when paired with its screenshotting capabilities.
But I do want to note that it doesn’t use any models under the hood (no VLMs/LLMs/even OCR models natively), and it’s not a replacement for VLM-based OCR solutions. It is fast because it is heuristic based! I attached a comparison table below. ✅ It is really good at text extraction and even table extraction, specifically for LLM understanding. It will lay the text out in a manner that’s easy for humans/AI to understand. ✅ It is great for assistant coding agents because the agent harness can use its text parsing to do a “fast” step, and then its screenshot capabilities to “dive deep” into a specific page 🚫 It is not great over scanned pages/visuals/anything requiring OCR. We do have OOB integrations with EasyOCR and PaddleOCR 🚫It doesn’t do layout detection and segmentation - it won’t draw bounding boxes over different elements on the page (though it does have word-level bounding boxes!)
Tl;dr it’s great for plugging into an AI assistant tool. If you’re trying to OCR a bunch of docs in batch, check out LlamaParse :)
LiteParse: https://t.co/JNER0mVcB8 LlamaParse: https://t.co/TqP6OT5U5O\n\nQT @jerryjliu0: Introducing LiteParse - the best model-free document parsing tool for AI agents 💫
✅ It’s completely open-source and free. ✅ No GPU required, will process ~500 pages in 2 seconds on commodity hardware ✅ More accurate than PyPDF, PyMuPDF, Markdown. Also way more readable - see below for how we parse tables!! ✅ Supports 50+ file formats, from PDFs to Office docs to images ✅ Is designed to plug and play with Claude Code, OpenClaw, and any other AI agent with a one-line skills install. Supports native screenshotting capabilities.
We spent years building up LlamaParse by orchestrating state-of-the-art VLMs over the most complex documents. Along the way we realized that you could get quite far on most docs through fast and cheap text parsing.
Take a look at the video below. For really complex tables within PDFs, we output them in a spatial grid that’s both AI and human-interpretable. Any other free/light parser light PyPDF will destroy the representation of this table and output a sequential list.
This is not a replacement for a VLM-based OCR tool (it requires 0 GPUs and doesn’t use models), but it is shocking how good it is to parse most documents.
Huge shoutout to @LoganMarkewich and @itsclelia for all the work here.
Come check it out: https://t.co/qmpDwlkidZ Repo: https://t.co/JNER0mVcB8
See 1 related tweets
- @itsclelia: RT @jerryjliu0: Last week we launched LiteParse - a free and fast document parser that provides more...
18. aakashgupta (Group Score: 58.8 | Individual: 32.4)
Cluster: 2 tweets | Engagement: 54 (Avg: 598) | Type: Tech
The single most expensive mistake I made testing Perplexity Computer:
Prompting it like a chatbot.
"Research my competitors" burned 800 credits. "Score these 5 competitors on pricing, positioning, and feature gaps using their public product pages" burned 200 credits for better output.
Computer spawns sub-agents for every ambiguous instruction. Vague prompt = more agents = more credits = worse results. The credit system punishes lazy prompting harder than any AI tool I've tested.
The five-rule Prompt Spec in this guide exists because I burned through my entire credit bonus learning this the hard way. Worth reading before you start a single task.\n\nQT @aakashgupta: For $20/month and zero setup, you can now run parallel AI agents that deliver finished work while you sleep.
Perplexity shipped Computer. Back on Ramp's fastest-growing B2B software list. 19+ AI models. 400+ connectors. The reason isn't search anymore.
Every take I've seen focuses on the "AI assistant" framing. They're all underselling it. Computer doesn't give you suggestions. It delivers the finished thing. Research reports with source citations. Deployed dashboards with shareable links. Cleaned datasets with charts. Launch kits with positioning docs and email drafts.
Three things make it different from everything else out there. Cloud execution, so your laptop can be closed. Parallel agents, so five tasks run simultaneously. And persistent memory, so you stop re-explaining yourself every session.
I pointed it at Notion's product pages. 28 pages scored across 5 criteria, competitive benchmarks against Coda and Slite, with specific recommendations per page. That's a $15K messaging audit. Took about 20 minutes.
But credits disappear fast if you don't know how to prompt it. I burned hundreds learning this. Built a five-rule Prompt Spec that cuts cost by 60%+.
I spent weeks testing it. Today's guide has the six PM use cases, exact prompts, the credit-saving system, and an honest comparison against Claude Code, Cowork, and OpenClaw.
Full guide: https://t.co/xHaRK91SEA
See 1 related tweets
- @rohanpaul_ai: RT @aakashgupta: The single most expensive mistake I made testing Perplexity Computer:
Prompting it...
19. svpino (Group Score: 57.2 | Individual: 29.6)
Cluster: 2 tweets | Engagement: 212 (Avg: 239) | Type: Tech
No shit! I’m sure nobody saw this coming.\n\nQT @GOrlanski: We found that agents generate progressively worse code with each iteration. Real developers do not.
SlopCodeBench is the only eval that faithfully measures quality degradation on iterative, long-horizon coding tasks.
https://t.co/JXGHC4w0bv https://t.co/RQkB8wdzAu 🧵 https://t.co/dOvNkrFv2c
See 1 related tweets
- @badlogicgames: have looked into it yet, but interesting\n\nQT @GOrlanski: We found that agents generate progressive...
20. alex_prompter (Group Score: 53.8 | Individual: 53.8)
Cluster: 1 tweets | Engagement: 686 (Avg: 68) | Type: Tech
RT @godofprompt: 🚨 BREAKING: Claude has a secret mode called "Aristotle First Principles Deconstructor."
It strips any complex problem down to its fundamental truths, eliminates every assumption you didn't know you were making, and rebuilds the solution from zero.
Aristotle invented this method 2,400 years ago. Now Claude runs it in 30 seconds.
Here's how to activate it: