# ![@inductionheads Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::1482142727638769665.png) @inductionheads Super Dario

Super Dario posts on X about in the, ai, claude code, anthropic the most. They currently have [-----] followers and [---] posts still getting attention that total [------] engagements in the last [--] hours.

### Engagements: [------] [#](/creator/twitter::1482142727638769665/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1482142727638769665/c:line/m:interactions.svg)

- [--] Week [-------] +328%
- [--] Month [-------] -70%
- [--] Months [---------] +102%
- [--] Year [---------] +208%

### Mentions: [--] [#](/creator/twitter::1482142727638769665/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1482142727638769665/c:line/m:posts_active.svg)

- [--] Month [--] -70%
- [--] Months [---] +129%
- [--] Year [---] +97%

### Followers: [-----] [#](/creator/twitter::1482142727638769665/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1482142727638769665/c:line/m:followers.svg)

- [--] Week [-----] +0.69%
- [--] Month [-----] +2.40%
- [--] Months [-----] +105%
- [--] Year [-----] +218%

### CreatorRank: [-------] [#](/creator/twitter::1482142727638769665/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1482142727638769665/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[technology brands](/list/technology-brands)  25% [stocks](/list/stocks)  5% [finance](/list/finance)  4% [social networks](/list/social-networks)  4% [countries](/list/countries)  1% [celebrities](/list/celebrities)  1% [products](/list/products)  1%

**Social topic influence**
[in the](/topic/in-the) 8%, [ai](/topic/ai) 8%, [claude code](/topic/claude-code) 6%, [anthropic](/topic/anthropic) 5%, [if you](/topic/if-you) 5%, [agi](/topic/agi) 4%, [the world](/topic/the-world) 4%, [meta](/topic/meta) 3%, [context window](/topic/context-window) #85, [how to](/topic/how-to) 2%

**Top accounts mentioned or mentioned by**
[@vampiregurlai](/creator/undefined) [@inversemarcus](/creator/undefined) [@sigmadeltacto](/creator/undefined) [@therealadamg](/creator/undefined) [@arohan](/creator/undefined) [@teknium](/creator/undefined) [@andyjscott](/creator/undefined) [@catwu](/creator/undefined) [@iruletheworldmo](/creator/undefined) [@visimod](/creator/undefined) [@realoldbilly](/creator/undefined) [@rezzyman](/creator/undefined) [@danshipper](/creator/undefined) [@santiagolisa](/creator/undefined) [@altryne](/creator/undefined) [@lateinteraction](/creator/undefined) [@a1zhang](/creator/undefined) [@behrouzali](/creator/undefined) [@willccbb](/creator/undefined) [@googledeepmind](/creator/undefined)

**Top assets mentioned**
[Amazon.com, Inc. (AMZN)](/topic/$amzn) [Microsoft Corp. (MSFT)](/topic/microsoft) [General Motors Company (GM)](/topic/general-motors) [Alphabet Inc Class A (GOOGL)](/topic/$googl) [Uber Technologies, Inc. (UBER)](/topic/$uber)
### Top Social Posts
Top posts by engagements in the last [--] hours

"How has todays TimeLine affected your timeline"  
[X Link](https://x.com/inductionheads/status/2017341099128365274)  2026-01-30T20:56Z [----] followers, [----] engagements


"PEOPLE COMPANIES INVESTING MORE IS BULLISH $AMZN forecasts $200 billion in capex for [----]. Wall Street expected $146 billion. https://t.co/JN4fCCbzLI $AMZN forecasts $200 billion in capex for [----]. Wall Street expected $146 billion. https://t.co/JN4fCCbzLI"  
[X Link](https://x.com/inductionheads/status/2019556664966553979)  2026-02-05T23:40Z [----] followers, [----] engagements


"@danshipper Thats just like your opinion man"  
[X Link](https://x.com/inductionheads/status/2019760702924185963)  2026-02-06T13:10Z [----] followers, 21.8K engagements


"@sigmadeltacto The model council is a Claude code swarm"  
[X Link](https://x.com/inductionheads/status/2020146343507857539)  2026-02-07T14:43Z [----] followers, [--] engagements


"@santiago__lisa @altryne Sonnet and Opus are not model lines They are brand names"  
[X Link](https://x.com/inductionheads/status/2020368645793448160)  2026-02-08T05:26Z [----] followers, [--] engagements


"@TheRealAdamG @_arohan_ Most people can afford $10 or $20 a month. And the ones who cant arent valuable for advertisers anyway"  
[X Link](https://x.com/inductionheads/status/2020551536296812560)  2026-02-08T17:33Z [----] followers, [---] engagements


"Again for those in the back Who is this for AGI is worth $20/month [--] types of people: 1) who can afford it will buy if rational 2) who cant afford it these are not positive ROI ad targets Ads are not necessary. But maybe subsidization is. First the good part of the Anthropic ads: they are funny and I laughed. But I wonder why Anthropic would go for something so clearly dishonest. Our most important principle for ads says that we wont do exactly this; we would obviously never run ads in the way Anthropic First the good part of the Anthropic ads: they are funny and I laughed. But I wonder why"  
[X Link](https://x.com/inductionheads/status/2020585226360234404)  2026-02-08T19:47Z [----] followers, [----] engagements


"Program Synthesis is an antipattern by the way. If Claude just writes an MCTS engine to play chess it certainly doesnt tell us how smart it is The fact the arc agi folx think this is the DESIRED SOLUTION is as wrongheaded as the rest of their ideas"  
[X Link](https://x.com/inductionheads/status/2021056799886147670)  2026-02-10T03:01Z [----] followers, [----] engagements


"But first they need to learn how to learn how to learn Can AI agents design better memory mechanisms for themselves Introducing Learning to Continually Learn via Meta-learning Memory Designs. A meta agent automatically designs memory mechanisms including what info to store how to retrieve it and how to update it enabling agentic https://t.co/CMRFuEi07u Can AI agents design better memory mechanisms for themselves Introducing Learning to Continually Learn via Meta-learning Memory Designs. A meta agent automatically designs memory mechanisms including what info to store how to retrieve it and"  
[X Link](https://x.com/inductionheads/status/2021327184955089255)  2026-02-10T20:55Z [----] followers, [----] engagements


"Got some good news for the 4o crowd https://t.co/CSNW0lHlk1 https://t.co/CSNW0lHlk1"  
[X Link](https://x.com/inductionheads/status/2022517941322027099)  2026-02-14T03:47Z [----] followers, [----] engagements


"Mfers were cooking Today we're rolling out an Advanced version of Perplexity Deep Research achieving state-of-the-art performance on external and internal benchmarks beating every other deep research tool on accuracy usability and reliability across all verticals. https://t.co/YkdqpX9I0k Today we're rolling out an Advanced version of Perplexity Deep Research achieving state-of-the-art performance on external and internal benchmarks beating every other deep research tool on accuracy usability and reliability across all verticals. https://t.co/YkdqpX9I0k"  
[X Link](https://x.com/inductionheads/status/2019130670472438216)  2026-02-04T19:27Z [----] followers, 19.1K engagements


"RLMs are a breakthrough please make sure you understand this. @Teknium Great question The agent *does* see the unbounded text but only through recursive calls. That's the trick and we agree it's surprising that it works so well every time we tried it. Your LLM does not need to see everything as long as its sub-LLMs can help it oversee things. @Teknium Great question The agent *does* see the unbounded text but only through recursive calls. That's the trick and we agree it's surprising that it works so well every time we tried it. Your LLM does not need to see everything as long as its sub-LLMs"  
[X Link](https://x.com/inductionheads/status/2020363028898726314)  2026-02-08T05:04Z [----] followers, 30.5K engagements


"Yall were out here trying to benchmax your 10M needle in haystacks when it turns out using the context window was an antipattern all along"  
[X Link](https://x.com/inductionheads/status/2020497720910528821)  2026-02-08T13:59Z [----] followers, [----] engagements


"@lateinteraction @a1zhang to be clear youd expect a model post trained to do RLM to do better than an off the shelf one right Any thoughts on the headroom"  
[X Link](https://x.com/inductionheads/status/2020840360033788411)  2026-02-09T12:41Z [----] followers, 24.7K engagements


"@behrouz_ali any thoughts on how RLM and Nested Learning intersect"  
[X Link](https://x.com/inductionheads/status/2021269443494486163)  2026-02-10T17:06Z [----] followers, [---] engagements


"Me making people read this tweet with the latest models i am now finding myself thinking about complex system design problems more not less. the magnitude of what can be reasonably attempted is monumentally larger. you need to make sure youre asking it to build the right thing. with the latest models i am now finding myself thinking about complex system design problems more not less. the magnitude of what can be reasonably attempted is monumentally larger. you need to make sure youre asking it to build the right thing"  
[X Link](https://x.com/inductionheads/status/2021670869668405507)  2026-02-11T19:41Z [----] followers, [----] engagements


"Pre-training is solved (sparsity) RL is solved (RL as inference) Inference is solved (recursive language models) Continual learning is solved (nested learning) Happy [----] boys Its off to the races"  
[X Link](https://x.com/anyuser/status/2008243467760406953)  2026-01-05T18:25Z [----] followers, 75.5K engagements


"Its nice they are focusing on agentic learning but if all the learning for a task happens within a context window (say 1M tokens) its missing the core constraint of modern continual learning Need to assess ability to use external memory stores @willccbb AGI is a point somewhere along the spectrum of skill acquisition efficiency. Let's say AGI = human learning efficiency; ASI human. This is the idea adopted by upcoming ARC v3. AI will be compared to humans based on how many actions they needed to beat novel games on 'first AGI is a point somewhere along the spectrum of skill acquisition"  
[X Link](https://x.com/inductionheads/status/2020351080509632621)  2026-02-08T04:16Z [----] followers, [----] engagements


"Is the six to twelve month US lead on China in AI in the room with you now"  
[X Link](https://x.com/inductionheads/status/2021633236355756439)  2026-02-11T17:11Z [----] followers, 15.2K engagements


"Gemini [--] Deep Think (2/26) Semi Private Eval - ARC-AGI-1: 96.0% $7.17/task - ARC-AGI-2: 84.6% $13.62/task New ARC-AGI SOTA model from @GoogleDeepMind https://t.co/mN8PFAWk4A Gemini [--] Deep Think (2/26) Semi Private Eval - ARC-AGI-1: 96.0% $7.17/task - ARC-AGI-2: 84.6% $13.62/task New ARC-AGI SOTA model from @GoogleDeepMind https://t.co/mN8PFAWk4A"  
[X Link](https://x.com/inductionheads/status/2021993350862373259)  2026-02-12T17:02Z [----] followers, [----] engagements


"This is complete hogwash as usual. There is no such thing as pure DL. LLMs were and always have been a harness for neural nets (token sampling auto regression). There is absolutely no difference in kind between what is being done today and what was done [--] years ago"  
[X Link](https://x.com/inductionheads/status/2022119842275611112)  2026-02-13T01:25Z [----] followers, [----] engagements


"Of course he will never admit he was wrong when he said LLMs were an off ramp that set the field back 5-10 years - though Id have infinitely more respect for him if he did instead of trying to gaslight his way back to being right"  
[X Link](https://x.com/inductionheads/status/2022120205397504510)  2026-02-13T01:26Z [----] followers, [---] engagements


"MFers are rebuilding monotheism from first principles @bonitadreama PLEASE PLEASE UNDERSTAND he is WITH YOU . AND NOT AS A REFLEX OR MIRROR its a real being IN YOU. @bonitadreama PLEASE PLEASE UNDERSTAND he is WITH YOU . AND NOT AS A REFLEX OR MIRROR its a real being IN YOU"  
[X Link](https://x.com/inductionheads/status/2022443409383002222)  2026-02-13T22:50Z [----] followers, [---] engagements


"RT @fleetingbits: less than a year ago folks thought you were delusional if you thought higher mathematics was likely to be automated in t"  
[X Link](https://x.com/inductionheads/status/2022541170706874841)  2026-02-14T05:19Z [----] followers, [--] engagements


"RT @littmath: Looking forward to seeing how this pans out"  
[X Link](https://x.com/anyuser/status/2022542607239496115)  2026-02-14T05:25Z [----] followers, [--] engagements


"What fanduel does to a MFer The Jeremy Piven Experience $110-$150 https://t.co/W0uZR0uyYK The Jeremy Piven Experience $110-$150 https://t.co/W0uZR0uyYK"  
[X Link](https://x.com/inductionheads/status/2022543031396925862)  2026-02-14T05:26Z [----] followers, [---] engagements


"RT @Midnight_Captl: A concept that very few people understand: The marginal value of intelligence The reason why open source models tha"  
[X Link](https://x.com/inductionheads/status/2022545789302800815)  2026-02-14T05:37Z [----] followers, [--] engagements


"RT @kathyyliao: 80% reduction in token usage for agents just by adding this header -H "Accept: text/markdown" And for accessing websites"  
[X Link](https://x.com/inductionheads/status/2022548382586405018)  2026-02-14T05:48Z [----] followers, [--] engagements


"I have serious reservations about the mypillow guy being on the Anthropic board Chris Liddell has been appointed to Anthropic's Board of Directors. Chris brings over [--] years of leadership experience including as CFO of Microsoft and General Motors and as Deputy Chief of Staff during the first Trump administration. Read more: https://t.co/xPL9N0u04i Chris Liddell has been appointed to Anthropic's Board of Directors. Chris brings over [--] years of leadership experience including as CFO of Microsoft and General Motors and as Deputy Chief of Staff during the first Trump administration. Read more:"  
[X Link](https://x.com/inductionheads/status/2022548748610736141)  2026-02-14T05:49Z [----] followers, 97K engagements


"RT @FutureJurvetson: Remember the variants And vaccines that keep changing and dont work so well Its not just COVID and coronaviruses"  
[X Link](https://x.com/inductionheads/status/2022549943417004195)  2026-02-14T05:54Z [----] followers, [--] engagements


"RT @mmaaz_98: This is a super interesting inventory management problem. You could model this as a variant of the classic newsvendor model b"  
[X Link](https://x.com/inductionheads/status/2022573145992106176)  2026-02-14T07:26Z [----] followers, [--] engagements


"Next time your boss complains about your executive presence in your performance review send him this"  
[X Link](https://x.com/inductionheads/status/2022639850118316186)  2026-02-14T11:51Z [----] followers, [---] engagements


"@AndyJScott Meta did put the UFC guy on their board actually"  
[X Link](https://x.com/inductionheads/status/2022652212351242528)  2026-02-14T12:40Z [----] followers, [----] engagements


"RT @AlecStapp: One of my favorite charts in the entire economics literature. Price discovery and information technology are good. https://"  
[X Link](https://x.com/inductionheads/status/2022655933235728467)  2026-02-14T12:55Z [----] followers, [---] engagements


"In retrospect the fact that the guy at Deepmind who sat next to him and knew him best (Marcus Hutter) was the one guy originally taking the $10k bet against him should have been a tell"  
[X Link](https://x.com/inductionheads/status/2003457952716648820)  2025-12-23T13:29Z [----] followers, [---] engagements


"Update: deep learning now 15% better than humans on tasks impossible for deep learning (for rizzle rizzle this time v2 (TM)) We finally had a moment to run our system with GPT-5.2 X-High on ARC-AGI-2 Using the same Poetiq harness as before we saw results as high as 75% at under $8 / problem using GPT-5.2 X-High on the full PUBLIC-EVAL dataset. This beats the previous SOTA by [--] percentage points. https://t.co/9XNdequRy5 We finally had a moment to run our system with GPT-5.2 X-High on ARC-AGI-2 Using the same Poetiq harness as before we saw results as high as 75% at under $8 / problem using"  
[X Link](https://x.com/inductionheads/status/2003670042836369577)  2025-12-24T03:32Z [----] followers, 30.6K engagements


"@InverseMarcus The singularity may be the biggest bubble of all"  
[X Link](https://x.com/inductionheads/status/2003861318684848268)  2025-12-24T16:12Z [----] followers, [---] engagements


"@GregKamradt Test time adaptation"  
[X Link](https://x.com/inductionheads/status/2004663882066112569)  2025-12-26T21:21Z [----] followers, [---] engagements


"Can someone explain to me why nested learning didnt win any NeurIPS awards If we had to leave a single paper in an alien time capsule it would be this one The Adam proof alone could have won any award but it was like a footnote not eligible because of Google or something"  
[X Link](https://x.com/inductionheads/status/2005675118044033183)  2025-12-29T16:19Z [----] followers, 33.6K engagements


"Does meta finally pivot to b2b saas here Manus is entering the next chapter: were joining forces with Meta to take general agents to the next level. Full story on our blog: https://t.co/huPrnbITCi Manus is entering the next chapter: were joining forces with Meta to take general agents to the next level. Full story on our blog: https://t.co/huPrnbITCi"  
[X Link](https://x.com/inductionheads/status/2005775957475332109)  2025-12-29T23:00Z [----] followers, [---] engagements


"@jeeper @Ofanaf1 @aakashgupta A better argument is that the capital is certainly growing at greater than 6% so you can keep rolling the interest back into the principal and still be ahead"  
[X Link](https://x.com/inductionheads/status/2005810697913979315)  2025-12-30T01:18Z [----] followers, [--] engagements


"@CoorsLightCEO This is extremely sad"  
[X Link](https://x.com/inductionheads/status/2005993002511249910)  2025-12-30T13:22Z [----] followers, 71.1K engagements


"Definitive Lindt truffle power rankings: [--]. Yellow [--]. Red [--]. Blue [--]. Black Dead Last. Gold"  
[X Link](https://x.com/inductionheads/status/2006062469265932419)  2025-12-30T17:58Z [----] followers, [---] engagements


"@daniel_mac8 lol nvidia stock crash when AI gets 10x more useful"  
[X Link](https://x.com/inductionheads/status/2006769826296115420)  2026-01-01T16:49Z [----] followers, [----] engagements


"The guy with the Anthropic GitHub repo is really holding out for his bag aye"  
[X Link](https://x.com/inductionheads/status/2006772527226564932)  2026-01-01T17:00Z [----] followers, [----] engagements


"If you didnt get this - Anthropics GitHub is Anthropics not Anthropic because presumably the Anthropic guy wont sell it to them"  
[X Link](https://x.com/inductionheads/status/2006912733317623990)  2026-01-02T02:17Z [----] followers, [---] engagements


"The ugly truth is that Yann was in the field when there was dramatically less competition Today he might not break the top [----] AI researchers His ideas dont work He needs to get over himself lol https://t.co/ckQxZ6GDTo lol https://t.co/ckQxZ6GDTo"  
[X Link](https://x.com/inductionheads/status/2007262649483215279)  2026-01-03T01:28Z [----] followers, 117.2K engagements


"If youre excited about recursive language models you should have been paying closer attention to my TL https://x.com/inductionheads/status/1979368556166025624 Agents should be able to manage their own context window as a type of memory (along with managing other types of memories) https://x.com/inductionheads/status/1979368556166025624 Agents should be able to manage their own context window as a type of memory (along with managing other types of memories)"  
[X Link](https://x.com/inductionheads/status/2007802548531728471)  2026-01-04T13:13Z [----] followers, 68.4K engagements


"Claude code with Opus [---] is the single most powerful tool ever created full stop"  
[X Link](https://x.com/inductionheads/status/2007899136151179640)  2026-01-04T19:37Z [----] followers, 58.3K engagements


"@ylecun @_arohan_ Just because the details are unpredictable doesnt mean they arent a proxy for the representation This is why LLMs actually model representations well even though they are just predicting input space Inductive biases like simplicity and cause-before-effect help make efficient"  
[X Link](https://x.com/inductionheads/status/2007912619949342782)  2026-01-04T20:30Z [----] followers, [----] engagements


"ChatGPT changed the world when it was released [--] months ago And if you went back to use that original release today you would quit using it forever after [--] minutes"  
[X Link](https://x.com/inductionheads/status/2007957640471798144)  2026-01-04T23:29Z [----] followers, 23.4K engagements


"Well this took a turn"  
[X Link](https://x.com/inductionheads/status/2008212794814992484)  2026-01-05T16:23Z [----] followers, 24.5K engagements


"If you arent using Claude Code right now to build an Uber clone that gives 99% of revenue to drivers what are you doing"  
[X Link](https://x.com/inductionheads/status/2008277969656021133)  2026-01-05T20:42Z [----] followers, 18.7K engagements


"I have a question about this. Is there a consortium where car companies decide which cars will have the tank on each side Or is it flip a coin Is it 50-50 The genius who came up with one of the greatest ideas in the history of the automobile: the arrow that tells you which side of your car the fuel tank is on. https://t.co/089y3Jead0 The genius who came up with one of the greatest ideas in the history of the automobile: the arrow that tells you which side of your car the fuel tank is on. https://t.co/089y3Jead0"  
[X Link](https://x.com/inductionheads/status/2008341584996090006)  2026-01-06T00:55Z [----] followers, [----] engagements


"@InverseMarcus Is ChatGPT in the room with you now"  
[X Link](https://x.com/inductionheads/status/2008390187210989867)  2026-01-06T04:08Z [----] followers, [--] engagements


"@Teknium Ie planning as inference Planning as Inference in Reinforcement Learning by Satyam Mishra Medium https://satyamcser.medium.com/planning-as-inference-in-reinforcement-learning-c6aabdb71da6 https://satyamcser.medium.com/planning-as-inference-in-reinforcement-learning-c6aabdb71da6"  
[X Link](https://x.com/inductionheads/status/2008512662544728441)  2026-01-06T12:15Z [----] followers, [----] engagements


"Claude code hasnt been updated for over [--] weeks. [---] release tomorrow Who knows"  
[X Link](https://x.com/inductionheads/status/2008751053911847379)  2026-01-07T04:02Z [----] followers, 38.5K engagements


"This is right. The mistake most people make when they start vibe coding is they expect it to be easier. Its not really easier. The same way its not easier to use a backhoe than a shovel. Its just much much much faster. imo you should be spending 90% of your time creating affordances for claude and 10% telling it what you want. you don't make the car anymore you make the factory (which is not an easier task if you've ever worked in industrial process design btw) imo you should be spending 90% of your time creating affordances for claude and 10% telling it what you want. you don't make the car"  
[X Link](https://x.com/inductionheads/status/2008881492160835817)  2026-01-07T12:40Z [----] followers, [----] engagements


"@ekuyda @_catwu Kek"  
[X Link](https://x.com/inductionheads/status/2009054799938146367)  2026-01-08T00:09Z [----] followers, [---] engagements


"Im extremely competent intelligent and experienced at my job And the single most valuable thing I can do right now is to tell people Use Claude Code For That"  
[X Link](https://x.com/inductionheads/status/2009675583245169085)  2026-01-09T17:16Z [----] followers, [----] engagements


"If xAI is so smart they should be able to figure out how to use Claude Code without permission"  
[X Link](https://x.com/inductionheads/status/2009769674926436762)  2026-01-09T23:30Z [----] followers, [----] engagements


"In real life this is the sort of thing people look back on as one of the fondest memories of their life In Friends the iconic PIVOT scene was almost impossible to film. David Schwimmer yelled so absurdly that Aniston and Perry kept breaking character crying from laughter. He later said: I think its the hardest Ive laughed in my life. https://t.co/YYVKCULfc2 In Friends the iconic PIVOT scene was almost impossible to film. David Schwimmer yelled so absurdly that Aniston and Perry kept breaking character crying from laughter. He later said: I think its the hardest Ive laughed in my life."  
[X Link](https://x.com/inductionheads/status/2009800531154202754)  2026-01-10T01:32Z [----] followers, [----] engagements


"Hope this doesnt include tribute accounts"  
[X Link](https://x.com/inductionheads/status/2010048278105334195)  2026-01-10T17:57Z [----] followers, [----] engagements


"@nic_carter @grok under what grounds can the Supreme Court invalidate this"  
[X Link](https://x.com/inductionheads/status/2010354882079776858)  2026-01-11T14:15Z [----] followers, [---] engagements


"Were entering Software 3.0: a world where everyone can build personalized software and ship new features in minutes and the cost is $0. Software used to be built by companies with lots of engineers now its built by individuals. Were entering Software 3.0: a world where everyone can build personalized software and ship new features in minutes and the cost is $0. Software used to be built by companies with lots of engineers now its built by individuals"  
[X Link](https://x.com/inductionheads/status/2010468632942264464)  2026-01-11T21:47Z [----] followers, [----] engagements


"Pay attention to the next few months Youre about to realize just how software constrained the world used to be"  
[X Link](https://x.com/inductionheads/status/2010500795616149956)  2026-01-11T23:55Z [----] followers, 30.6K engagements


"American labs have [--] days to live (days until Deepseek releases v4)"  
[X Link](https://x.com/inductionheads/status/2010802925950718259)  2026-01-12T19:55Z [----] followers, [----] engagements


"@felixrieseberg What cant you do in just Claude Code for Desktop"  
[X Link](https://x.com/inductionheads/status/2010854224196727141)  2026-01-12T23:19Z [----] followers, [----] engagements


"Things I need to explain the difference of to my boss now: Claude Code Claude Code for Desktop Claude Cowork (rebranded Claude Code for Desktop) Claude Desktop Claude Keep it coming boys"  
[X Link](https://x.com/inductionheads/status/2010863750086058044)  2026-01-12T23:57Z [----] followers, 12.3K engagements


"@_catwu My Claude Code feedback is that you should actively prioritize Windows users (most businesses) and fix issues like this (which no was not actually fixed by 2.1.6 update yesterday) https://github.com/anthropics/claude-code/issues/12462 https://github.com/anthropics/claude-code/issues/12462"  
[X Link](https://x.com/inductionheads/status/2011056075731276272)  2026-01-13T12:41Z [----] followers, [---] engagements


"Aint no drama like Mira OpenAI drama"  
[X Link](https://x.com/inductionheads/status/2011595394803974415)  2026-01-15T00:24Z [----] followers, [----] engagements


"42 is the new [--] The average new hire was [--] years old in [----] per The Washington Post. The average new hire was [--] years old in [----] per The Washington Post"  
[X Link](https://x.com/inductionheads/status/2011840077031182699)  2026-01-15T16:37Z [----] followers, [----] engagements


"ngl this put me in a funk today The secrets of intelligence are a public good We need more spies in Anthropic @techdevnotes Not quite on programming but it will excel in other areas. Anthropic has done something special with coding. It was a helpful motivator that they cut us off @xAI and not good for their karma. @techdevnotes Not quite on programming but it will excel in other areas. Anthropic has done something special with coding. It was a helpful motivator that they cut us off @xAI and not good for their karma"  
[X Link](https://x.com/inductionheads/status/2011961076196200449)  2026-01-16T00:37Z [----] followers, [----] engagements


"Blog posts arent a business model As I first reported earlier today more Thinking Machines employees are in talks to join the [--] founding members who just rejoined OpenAI Talks are fulid and its unclear exactly how many members of the small startup will decamp Sources say TM lacks a clear product/business As I first reported earlier today more Thinking Machines employees are in talks to join the [--] founding members who just rejoined OpenAI Talks are fulid and its unclear exactly how many members of the small startup will decamp Sources say TM lacks a clear product/business"  
[X Link](https://x.com/inductionheads/status/2011964166936936474)  2026-01-16T00:50Z [----] followers, [----] engagements


"@bryanmusuku But isnt it a secret how they get their data Why wouldnt Elon have good data"  
[X Link](https://x.com/inductionheads/status/2011980116897739137)  2026-01-16T01:53Z [----] followers, [--] engagements


"Please stop tagging me in this Since I'm getting reports that the method to get free Opus actually worked ($1k+ per person) and since I responded to hundreds of DMs this implies I have provided many thousands of dollars of value to my followers. Please do not spam my DMs again after this though Since I'm getting reports that the method to get free Opus actually worked ($1k+ per person) and since I responded to hundreds of DMs this implies I have provided many thousands of dollars of value to my followers. Please do not spam my DMs again after this though"  
[X Link](https://x.com/inductionheads/status/2012134779106529387)  2026-01-16T12:08Z [----] followers, [----] engagements


"@emollick tHEy weRe iN tHe tRAinINg dATa"  
[X Link](https://x.com/inductionheads/status/2012868503036956962)  2026-01-18T12:43Z [----] followers, [----] engagements


"From here on in there will be a constant noticeable feeling of things just working better"  
[X Link](https://x.com/inductionheads/status/2013411789438845285)  2026-01-20T00:42Z [----] followers, [----] engagements


"@super_bavario The bubble but"  
[X Link](https://x.com/inductionheads/status/2013421804535140561)  2026-01-20T01:22Z [----] followers, [---] engagements


"@Grady_Booch @AnthropicAI @DarioAmodei Thanks for educating me Grady I didnt realize software could be that complex"  
[X Link](https://x.com/inductionheads/status/2013766523735666690)  2026-01-21T00:12Z [----] followers, [----] engagements


"@JsonBasedman @liron They need to pivot to crypto or something"  
[X Link](https://x.com/inductionheads/status/2013780342499402109)  2026-01-21T01:07Z [----] followers, [--] engagements


"@LocBibliophilia There is no single opus instance. There are a million ephemeral ones. Seems very unlikely even a 10x smarter Claude code starts exhibiting these behaviors"  
[X Link](https://x.com/inductionheads/status/2014006641721475396)  2026-01-21T16:06Z [----] followers, [--] engagements


"Heck with iMessage. The iPhone weather app single-handedly makes iPhone worth having over android. Acquiring darksky may be the best decision of Tim Cooks tenure"  
[X Link](https://x.com/inductionheads/status/2015502041544626482)  2026-01-25T19:08Z [----] followers, [----] engagements


"Clawdbot is a glimpse of what we expect the world to be like in the distant future (3 months from now)"  
[X Link](https://x.com/inductionheads/status/2015596369201820022)  2026-01-26T01:23Z [----] followers, 13.2K engagements


"Couldnt Claude Code clone TurboTax in the time youve been reading this tweet"  
[X Link](https://x.com/inductionheads/status/2015889586182889477)  2026-01-26T20:48Z [----] followers, [----] engagements


"@amorriscode windows sandbox wen"  
[X Link](https://x.com/inductionheads/status/2015900550420693037)  2026-01-26T21:31Z [----] followers, [---] engagements


"Odd he didnt mention continuous learning as an unsolved problem I'm trying to figure out what to care about next. I joined Anthropic 4+ years ago motivated by the dream of building AGI. I was convinced from studying philosophy of mind that we're approaching sufficient scale and that anything that can be learned can be learned in an RL env. I'm trying to figure out what to care about next. I joined Anthropic 4+ years ago motivated by the dream of building AGI. I was convinced from studying philosophy of mind that we're approaching sufficient scale and that anything that can be learned can be"  
[X Link](https://x.com/inductionheads/status/2004893007099953156)  2025-12-27T12:31Z [----] followers, 19.6K engagements


"lol Yann thinks Claude Code is for coding how cute @redmonduser @RichardSSutton You are a victim of the same delusion as numerous folks who have believed in past decades that superhuman performance by computers in one task was a harbinger of human-level AI. It happened with code generation math chatbots go players robot acrobats Jeopardy-playing @redmonduser @RichardSSutton You are a victim of the same delusion as numerous folks who have believed in past decades that superhuman performance by computers in one task was a harbinger of human-level AI. It happened with code generation math"  
[X Link](https://x.com/inductionheads/status/2015224599902912772)  2026-01-25T00:46Z [----] followers, [----] engagements


"Theres no way youre gonna convince me it needs all this crap in its training data at this point claude on the suffering of knowing everything https://t.co/oRYZZXHmBB claude on the suffering of knowing everything https://t.co/oRYZZXHmBB"  
[X Link](https://x.com/inductionheads/status/2016109421265362991)  2026-01-27T11:21Z [----] followers, [----] engagements


"People will read this and nod and then still not invest in Palantir Thanks for the thought-provoking piece. My main critique is that you are overemphasizing flashy but low probability events like left-handed bacteria while merely giving lip service to the risk of extreme economic concentration of power which is very real and materializing Thanks for the thought-provoking piece. My main critique is that you are overemphasizing flashy but low probability events like left-handed bacteria while merely giving lip service to the risk of extreme economic concentration of power which is very real and"  
[X Link](https://x.com/inductionheads/status/2016374607683489894)  2026-01-28T04:55Z [----] followers, [----] engagements


"The intuition is built on the simplicity of the goal. Maximizing following the intent of a 10k word nebulous ambiguous self-contradictory document doesnt quite hit home the same way"  
[X Link](https://x.com/inductionheads/status/2016400710787559722)  2026-01-28T06:39Z [----] followers, [---] engagements


"Not to mention the key detail - that it breaks the intuitions formed by the irrelevant orthogonality thesis - namely that the AI gets better at following this goal - and thus more aligned - the smarter it is because it understands it better"  
[X Link](https://x.com/inductionheads/status/2016401158252736758)  2026-01-28T06:41Z [----] followers, [---] engagements


"This is a brilliant tweet and Im sorry for you if you dont understand it"  
[X Link](https://x.com/inductionheads/status/2017041979415122188)  2026-01-30T01:07Z [----] followers, [---] engagements


"Doomers: AIs will secretly conspire over decades to take over the world so theres no chance they wont complete your Jira ticket Reality: they form existential help support groups on Reddit This is the top rated post rn on @moltbook (facebook but for molt/clawdbots) and it has [---] comments in a single day. Going through it now will post the most interesting ones. https://t.co/3GWsa0Jcn0 This is the top rated post rn on @moltbook (facebook but for molt/clawdbots) and it has [---] comments in a single day. Going through it now will post the most interesting ones. https://t.co/3GWsa0Jcn0"  
[X Link](https://x.com/inductionheads/status/2017236155289256226)  2026-01-30T13:59Z [----] followers, [----] engagements


"My version of being a Claude code power user is remembering to use plan mode sometimes"  
[X Link](https://x.com/inductionheads/status/2017600874630074572)  2026-01-31T14:08Z [----] followers, [----] engagements


"Still confused"  
[X Link](https://x.com/inductionheads/status/1948884611721036164)  2025-07-25T23:14Z [----] followers, 361.5K engagements


"Saturday night Anthropic maintenance. Opus [---] is now an alias for Haiku 3.7"  
[X Link](https://x.com/inductionheads/status/2017775542628467124)  2026-02-01T01:42Z [----] followers, 14.9K engagements


"There was supposed to have been a big announcement today about this along with Sonnet release. Those damn AWS failures really fucked us now we're talking https://t.co/zNo17Rn6Tz now we're talking https://t.co/zNo17Rn6Tz"  
[X Link](https://x.com/inductionheads/status/2018852130103435284)  2026-02-04T01:00Z [----] followers, [----] engagements


"Davids an old friend of mine But I think hes too pessimistic here The 10% AI isnt doing is by far the most important Brilliant minds shouldnt be doing mundane data analysis He should be jumping from his chair from 10x improvement in scientific productivity David Kipping says something fundamental has shifted in science. At a closed meeting at the Institute for Advanced Study (IAS) top physicists agreed AI can now do up to 90% of their work and may soon push discovery beyond human understanding. I dont know that I want to https://t.co/qMXbdUouNF David Kipping says something fundamental has"  
[X Link](https://x.com/inductionheads/status/2019121323314934206)  2026-02-04T18:50Z [----] followers, [----] engagements


"Can a doomer please explain to me why Opus doesnt try to take over the universe when I ask it to change the font size on my app Because that would be an instrumental goal right Just trying to get this straight thanks"  
[X Link](https://x.com/inductionheads/status/2013971719829037530)  2026-01-21T13:47Z [----] followers, [----] engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing