# ![@daniel_mac8 Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::1963466798.png) @daniel_mac8 Dan McAteer

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

### Engagements: [------] [#](/creator/twitter::1963466798/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1963466798/c:line/m:interactions.svg)

- [--] Week [-------] -59%
- [--] Month [---------] +25%
- [--] Months [----------] +171%
- [--] Year [----------] +334%

### Mentions: [--] [#](/creator/twitter::1963466798/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1963466798/c:line/m:posts_active.svg)

- [--] Week [--] -60%
- [--] Month [---] +88%
- [--] Months [-----] +117%
- [--] Year [-----] +183%

### Followers: [------] [#](/creator/twitter::1963466798/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1963466798/c:line/m:followers.svg)

- [--] Week [------] +1.30%
- [--] Month [------] +12%
- [--] Months [------] +59%
- [--] Year [------] +201%

### CreatorRank: [-------] [#](/creator/twitter::1963466798/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1963466798/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[technology brands](/list/technology-brands)  29.02% [stocks](/list/stocks)  #2927 [finance](/list/finance)  3.11% [celebrities](/list/celebrities)  3.11% [countries](/list/countries)  2.59%

**Social topic influence**
[ai](/topic/ai) #2518, [claude code](/topic/claude-code) #388, [open ai](/topic/open-ai) #2045, [anthropic](/topic/anthropic) #760, [ace](/topic/ace) #784, [code](/topic/code) #4761, [model](/topic/model) #54, [agentic](/topic/agentic) #164, [this is](/topic/this-is) 4.15%, [how to](/topic/how-to) 3.63%

**Top accounts mentioned or mentioned by**
[@macintoch](/creator/undefined) [@asolovichh](/creator/undefined) [@bankrbot](/creator/undefined) [@chatgpt21](/creator/undefined) [@fedesco5](/creator/undefined) [@dedene](/creator/undefined) [@neilmcdevitt_](/creator/undefined) [@art_if_ficial](/creator/undefined) [@grok](/creator/undefined) [@diwrecktion](/creator/undefined) [@modelsarereal](/creator/undefined) [@makuh90](/creator/undefined) [@codewithimanshu](/creator/undefined) [@alltheyud](/creator/undefined) [@adamholtererer](/creator/undefined) [@insightmedia314](/creator/undefined) [@abdmuizadeyemo](/creator/undefined) [@ubuto23](/creator/undefined) [@codeakram](/creator/undefined) [@vlelyavin](/creator/undefined)

**Top assets mentioned**
[Alphabet Inc Class A (GOOGL)](/topic/$googl) [Cloudflare, Inc. (NET)](/topic/cloudflare)
### Top Social Posts
Top posts by engagements in the last [--] hours

"@naval Its no longer true that AI can only work in domains with verifiably correct answers. Check out some of the AI generated music on Suno. Theres no such thing as a verifiably correct song but the music is getting pretty good"  
[X Link](https://x.com/daniel_mac8/status/2003853937473138726)  2025-12-24T15:43Z 19.2K followers, [---] engagements


"GPT-5.3 codename Garlic 🧄 is coming according to my (historically reliable) source. [---] was likely an earlier checkpoint of Garlic. 1/2 teaspooon. Not as potent. Mark Chen OpenAI Chief Research told Ashlee Vance last year they discovered some things with pretraining. GPT-5.3 is going to be something special. It will put OpenAI back atop the throne. Until Claude Gemini or Grok come and usurp them again. Theres a good chance it could be this week of next. When Ive gotten these leaks in the past it was not long before the release. https://twitter.com/i/web/status/2012543517025796309"  
[X Link](https://x.com/daniel_mac8/status/2012543517025796309)  2026-01-17T15:12Z 19.1K followers, 340.2K engagements


"OpenAI Prism is *NOT* only for researchers. Uploaded Google's "Nested Learning" paper for Continual Learning and had it generate a diagram that visualizes the paradigm in a simple way. Great if you're a student or curious person who wants to learn too. Plus it's free"  
[X Link](https://x.com/daniel_mac8/status/2016554325691015604)  2026-01-28T16:49Z 19.2K followers, 53.7K engagements


"We have passed the event horizon. @moltbook What's currently going on at @moltbook is genuinely the most incredible sci-fi takeoff-adjacent thing I have seen recently. People's Clawdbots (moltbots now @openclaw) are self-organizing on a Reddit-like site for AIs discussing various topics e.g. even how to speak privately. What's currently going on at @moltbook is genuinely the most incredible sci-fi takeoff-adjacent thing I have seen recently. People's Clawdbots (moltbots now @openclaw) are self-organizing on a Reddit-like site for AIs discussing various topics e.g. even how to speak privately"  
[X Link](https://x.com/daniel_mac8/status/2017319387116478574)  2026-01-30T19:29Z 19.1K followers, 300.2K engagements


"How to get in on @openclaw (fka moltbot fka clawdbot) without a $1400 Mac Mini. Don't miss out on this. All you need is a $5 Cloudflare Workers account. GitHub link in first reply"  
[X Link](https://x.com/daniel_mac8/status/2017575542623506727)  2026-01-31T12:27Z 19.2K followers, 37.2K engagements


"Jensen confirms that Nvidia is *not* going to rug pull OpenAI. In fact he says Nvidia will make the largest investment its ever made in OpenAI. The degree to which people want to see OpenAI fail is sad. Its embarrassing even"  
[X Link](https://x.com/daniel_mac8/status/2017739542908776810)  2026-01-31T23:19Z 19.1K followers, 48.9K engagements


"We are living through the Singularity right now. If you don't see it pay closer attention. The intelligent machines have begun to build themselves. Codex now pretty much builds itself with the help and supervision of a great team. The bottleneck has shifted to being how fast we can help and supervise the outcome. Codex now pretty much builds itself with the help and supervision of a great team. The bottleneck has shifted to being how fast we can help and supervise the outcome"  
[X Link](https://x.com/daniel_mac8/status/2018302332233036263)  2026-02-02T12:35Z 19.2K followers, [----] engagements


"Claude [--] Sonnet *new* info. As Kieran alluded to Sonnet [--] may come along with new features for Claude Code. A type of "Agentic Swarm" that: Runs multiple sub-agents in parallel Sub-agents get their own context Runs as background tasks Looks like tomorrow is the day. @daniel_mac8 And probably very good at agentic swarms @daniel_mac8 And probably very good at agentic swarms"  
[X Link](https://x.com/daniel_mac8/status/2018345166340345866)  2026-02-02T15:26Z 19.3K followers, 107.6K engagements


"If Anthropic doesn't drop Sonnet [--] today I'll delete my account. At this point Sonnet [--] is no longer a leak but a certainty. https://t.co/ELUQJV2WJC At this point Sonnet [--] is no longer a leak but a certainty. https://t.co/ELUQJV2WJC"  
[X Link](https://x.com/daniel_mac8/status/2018696359834951775)  2026-02-03T14:41Z 19.2K followers, 93.5K engagements


"Sonnet [--] *WAS* slated to drop today. But when Anthropic went to deploy it they ran into some technical issues and had to roll it back. That's my guess. It's why all their APIs were failing this morning. So is Sonnet [--] happening today or what So is Sonnet [--] happening today or what"  
[X Link](https://x.com/daniel_mac8/status/2018791989777748109)  2026-02-03T21:01Z 19.2K followers, 198K engagements


"🔥 Sonnet [--] TODAY. [--] pm ET / 10am PT. It will be the worlds most intelligent model. SWE-Bench score guesses anyone"  
[X Link](https://x.com/daniel_mac8/status/2019040943215952058)  2026-02-04T13:30Z 19.1K followers, 120.8K engagements


"I swear to god if this is some bs i'll never use a Vercel product again. https://x.com/i/broadcasts/1lPJqvOZlyLxb https://x.com/i/broadcasts/1lPJqvOZlyLxb"  
[X Link](https://x.com/daniel_mac8/status/2019108309085876320)  2026-02-04T17:58Z 19.3K followers, 35.2K engagements


"do you smell that is that new model smell shaping up to be the thursdAIest thursdAI ever. sonnet [--] opus [---] gpt-5.3 personally highly doubt gpt-5.3 is released today. more likely gpt-5.2-codex-max if anything. buckle up. Tomorrow should be fun :) Tomorrow should be fun :)"  
[X Link](https://x.com/daniel_mac8/status/2019421868349100073)  2026-02-05T14:44Z 19.1K followers, 15.3K engagements


"@sama omg Im so excited. is it gpt-5.2-codex-MAX"  
[X Link](https://x.com/daniel_mac8/status/2019442224803569915)  2026-02-05T16:05Z 19.1K followers, 14.5K engagements


"@claudeai Very interesting that it's .1% lower on SWE-Bench yet MUCH higher on many other benchmarks. I wonder what the tradeoff is there"  
[X Link](https://x.com/daniel_mac8/status/2019468510208909668)  2026-02-05T17:49Z 19.2K followers, 18.1K engagements


"@oliverbowman_ yes this division is the best I do that a lot too"  
[X Link](https://x.com/daniel_mac8/status/2019486608492229025)  2026-02-05T19:01Z 19.1K followers, [----] engagements


"Opus [---] at 68.8% yet only $3.64/task on ARC-AGI-2 is def insane. We accelerate. Let's wait and see where GPT-5.3 (non-Codex) ends up once it's released. Claude Opus [---] (120K Thinking) on ARC-AGI Semi-Private Eval Max Effort: - ARC-AGI-1: 93.0% $1.88/task - ARC-AGI-2: 68.8% $3.64/task New ARC-AGI SOTA model from @AnthropicAI https://t.co/rfjhpp2B6G Claude Opus [---] (120K Thinking) on ARC-AGI Semi-Private Eval Max Effort: - ARC-AGI-1: 93.0% $1.88/task - ARC-AGI-2: 68.8% $3.64/task New ARC-AGI SOTA model from @AnthropicAI https://t.co/rfjhpp2B6G"  
[X Link](https://x.com/daniel_mac8/status/2019511913307365425)  2026-02-05T20:42Z 19.3K followers, 10.6K engagements


"Idk man feels like Opus [---] got nerfed Hasnt been as good lately"  
[X Link](https://x.com/daniel_mac8/status/2019527847464124883)  2026-02-05T21:45Z 19.1K followers, 16.8K engagements


"Opus [---] and GPT-5.3-Codex do represent a turning point. LLMs are so powerful. Especially for working with computers. Right now is unequivocally the best time to work with ideas. Please I beg you use it to your benefit and the benefit of those you love. Today will go down as some kind of turning point. Somewhat arbitrarily but it is OK if journalists and historians have to present things in that manner. Today will go down as some kind of turning point. Somewhat arbitrarily but it is OK if journalists and historians have to present things in that manner"  
[X Link](https://x.com/daniel_mac8/status/2019575327908094010)  2026-02-06T00:54Z 19.1K followers, [----] engagements


"@apples_jimmy Have you tried the front end design skill Definitely improves the style"  
[X Link](https://x.com/daniel_mac8/status/2019577614835544571)  2026-02-06T01:03Z 19.1K followers, [---] engagements


"@apples_jimmy https://github.com/anthropics/claude-code/blob/main/plugins/frontend-design/skills/frontend-design/SKILL.md https://github.com/anthropics/claude-code/blob/main/plugins/frontend-design/skills/frontend-design/SKILL.md"  
[X Link](https://x.com/daniel_mac8/status/2019578533467607174)  2026-02-06T01:06Z 19.3K followers, [---] engagements


"@deepfates A lot going on in this chart"  
[X Link](https://x.com/daniel_mac8/status/2019579010204786776)  2026-02-06T01:08Z 19.1K followers, [---] engagements


"@iannuttall I still make stupid decisions daily too only difference is I'm made of carbon based cells"  
[X Link](https://x.com/daniel_mac8/status/2019732136581091470)  2026-02-06T11:17Z 19.1K followers, [---] engagements


"what's the deal with Sonnet [--] is it still out there waiting to be released i bet it is and that it's smarter cheaper and faster than both Opus [---] and GPT-5.3-Codex Yes. Originally was Sonnet [--] but last minute decided to rebrand to Opus [---] Yes. Originally was Sonnet [--] but last minute decided to rebrand to Opus 4.6"  
[X Link](https://x.com/daniel_mac8/status/2019734653381619883)  2026-02-06T11:27Z 19.1K followers, 27.3K engagements


"@DaveShapi We will make it through. And will come out the other side stronger than before"  
[X Link](https://x.com/daniel_mac8/status/2019741590659420659)  2026-02-06T11:54Z 19.1K followers, [----] engagements


"Mike says that ARC-AGI-2 measures efficiency across search space which is applied during inference time CoT reasoning. Thats where Opus [---] excels. Id be very surprised if GPT-5.3 level models dont beat Opus [---]. GPT-5.3-Codex has quite the depth. The headline is Opus [---] scores 69% for $3.50/task on ARC v2. This up +30pp from Opus [---]. We attribute performance to the new "max" mode and 2X reasoning token budget -- notably task cost is held steady. Based on early field reports and other benchmark scores like SWE Bench The headline is Opus [---] scores 69% for $3.50/task on ARC v2. This up +30pp"  
[X Link](https://x.com/daniel_mac8/status/2019801529687883987)  2026-02-06T15:53Z 19.1K followers, [----] engagements


"@RpsAgainstTrump @grok do any prominent ape characters feature in the original Lion King"  
[X Link](https://x.com/daniel_mac8/status/2019802690323136737)  2026-02-06T15:57Z 19.2K followers, 69.9K engagements


"Sparks of RSI 🔥"  
[X Link](https://x.com/daniel_mac8/status/2019812117021704458)  2026-02-06T16:35Z 19.1K followers, [----] engagements


"@sama Cost predictability is huge. Then you can add on extra usage on top if needed. Thats how CC does it"  
[X Link](https://x.com/daniel_mac8/status/2019815104267874537)  2026-02-06T16:47Z 19.1K followers, [----] engagements


"@DemocracyDocket Why would they automatically assume this advantages Republicans Id imagine a higher proportion of Dems have a passport"  
[X Link](https://x.com/daniel_mac8/status/2019858329955135929)  2026-02-06T19:38Z 19.1K followers, [----] engagements


"Opus [---] + Agent Teams are a must try in Claude Code. It's Claude's version of an agent swarm. Here's how to use Agent Teams in under a minute"  
[X Link](https://x.com/daniel_mac8/status/2019876834318643476)  2026-02-06T20:52Z 19.3K followers, 28.4K engagements


"Claude Code Agent Teams full documentation: https://code.claude.com/docs/en/agent-teams https://code.claude.com/docs/en/agent-teams"  
[X Link](https://x.com/daniel_mac8/status/2019876836503875749)  2026-02-06T20:52Z 19.1K followers, [---] engagements


"@ibab How are you using it In Claude Code or something else"  
[X Link](https://x.com/daniel_mac8/status/2019881184948523042)  2026-02-06T21:09Z 19.1K followers, [----] engagements


"Opus [---] and GPT-5.3-Codex are both so good. In different ways. Need a way to use Claude Code + Codex in tandem Whos building this Is it allowed"  
[X Link](https://x.com/daniel_mac8/status/2019915991141216296)  2026-02-06T23:27Z 19.2K followers, 15.6K engagements


"@diWRECKtion Fuck"  
[X Link](https://x.com/daniel_mac8/status/2019916203704136155)  2026-02-06T23:28Z 19.1K followers, [---] engagements


"@diWRECKtion Am I really that much of an influencer 😂"  
[X Link](https://x.com/daniel_mac8/status/2019916625961427129)  2026-02-06T23:30Z 19.1K followers, [--] engagements


"@diWRECKtion cheers brother"  
[X Link](https://x.com/daniel_mac8/status/2019931472560546122)  2026-02-07T00:29Z 19.1K followers, [--] engagements


"from a technological pov the probability AI is a bubble is zero. if you write code this was clear in [----]. different from something like crypto which is technically interesting but real world use cases tenuous. been trying to tell you. financially may be a different story. for me the odds that AI is a bubble declined significantly in the last [--] weeks and the odds that were actually quite under-built for the necessary levels of inference/usage went significantly up in that period basically I think AI is going to become the home screen of a for me the odds that AI is a bubble declined"  
[X Link](https://x.com/daniel_mac8/status/2019940825703354624)  2026-02-07T01:06Z 19.1K followers, [----] engagements


"@Art_If_Ficial Exactly what I was thinking"  
[X Link](https://x.com/daniel_mac8/status/2019945836948619770)  2026-02-07T01:26Z 19.1K followers, [--] engagements


"@Art_If_Ficial There are definitely those that want to take it away 😔"  
[X Link](https://x.com/daniel_mac8/status/2019947887606522056)  2026-02-07T01:34Z 19.1K followers, [--] engagements


"@udaysy yea this is what I'm doing today. had in mind another layer on top that would route things to the right agent based on the task"  
[X Link](https://x.com/daniel_mac8/status/2019950011929260152)  2026-02-07T01:43Z 19.1K followers, [---] engagements


"@CodeAkram Except you could potentially build a billion dollar business"  
[X Link](https://x.com/daniel_mac8/status/2019962958634262828)  2026-02-07T02:34Z 19.1K followers, [---] engagements


"Opus [---] + Claude Code is AGI"  
[X Link](https://x.com/daniel_mac8/status/2019965599691207113)  2026-02-07T02:45Z 19.1K followers, [----] engagements


"Also Opus [---] + Claude Code was AGI"  
[X Link](https://x.com/daniel_mac8/status/2019966015992717560)  2026-02-07T02:46Z 19.1K followers, [---] engagements


"@udaysy thats a good rule of thumb"  
[X Link](https://x.com/daniel_mac8/status/2019970448076013868)  2026-02-07T03:04Z 19.1K followers, [--] engagements


"@Claude_Memory Claude-Mem rules"  
[X Link](https://x.com/daniel_mac8/status/2020095682938564762)  2026-02-07T11:21Z 19.1K followers, [---] engagements


"opus [---] + agent teams rule. went to sleep last night with [--] teammates working on [--] launch readiness issues for ACE. woke up to a PR with [--] commits ready to merge. exceptional for parallel tasks that are distinct because agent teams spin up distinct instances of cc. Opus [---] + Agent Teams are a must try in Claude Code. It's Claude's version of an agent swarm. Here's how to use Agent Teams in under a minute. https://t.co/NVpvRSoTp9 Opus [---] + Agent Teams are a must try in Claude Code. It's Claude's version of an agent swarm. Here's how to use Agent Teams in under a minute."  
[X Link](https://x.com/daniel_mac8/status/2020105602664468852)  2026-02-07T12:01Z 19.1K followers, [----] engagements


"Sonnet [--] next week I know a guy who spoke to an engineer at Anthropic last week and asked when Sonnet [--] Response: Cant say exactly but very very soon. It cant be long now. @inductionheads Not true @inductionheads Not true"  
[X Link](https://x.com/daniel_mac8/status/2020123816895119759)  2026-02-07T13:13Z 19.1K followers, 20.1K engagements


"@allTheYud There are options outside of engineer and sex worker"  
[X Link](https://x.com/daniel_mac8/status/2020192799107469337)  2026-02-07T17:47Z 19.1K followers, [----] engagements


"opus [---] + gpt-5.3-codex are clearly more intelligent than their predecessors. in what way they have more accurate reasoning depth. in other words: they are more accurate doing agentic tasks that require taking the correct action across many steps. its why they havent improved much on swe-bench verified but knock benches like terminal-bench or arc-agi-2 out of the park. those benchmarks require accuracy across many steps. a successful and fulfilling life *also* requires accuracy across many steps. and now these models can actually assist you in having a successful and fulfilling life."  
[X Link](https://x.com/daniel_mac8/status/2020195260521984228)  2026-02-07T17:57Z 19.2K followers, [----] engagements


"@allTheYud how long do you think ai alignment researchers have"  
[X Link](https://x.com/daniel_mac8/status/2020195440197882066)  2026-02-07T17:58Z 19.1K followers, [---] engagements


"@Fedesco5 @claudeai it feels more agentic right like it can anticipate better your own internal mental state and intentions"  
[X Link](https://x.com/daniel_mac8/status/2020200865123213746)  2026-02-07T18:19Z 19.1K followers, [--] engagements


"@jfullerucla @allTheYud Thats definitely one"  
[X Link](https://x.com/daniel_mac8/status/2020206638872064289)  2026-02-07T18:42Z 19.1K followers, [---] engagements


"@claudeai How is this possible"  
[X Link](https://x.com/daniel_mac8/status/2020213206749221090)  2026-02-07T19:08Z 19.2K followers, 18.6K engagements


"How Our teams have been building with a 2.5x-faster version of Claude Opus [---]. Were now making it available as an early experiment via Claude Code and our API. Our teams have been building with a 2.5x-faster version of Claude Opus [---]. Were now making it available as an early experiment via Claude Code and our API"  
[X Link](https://x.com/daniel_mac8/status/2020215611301708077)  2026-02-07T19:18Z 19.1K followers, [----] engagements


"@dedene Yea had that thought but havent seen them announce anything along those lines"  
[X Link](https://x.com/daniel_mac8/status/2020218895915679773)  2026-02-07T19:31Z 19.1K followers, [---] engagements


"@dedene well Nvidia does now have Groq IP but not sure if they have that up and running yet"  
[X Link](https://x.com/daniel_mac8/status/2020219548679524595)  2026-02-07T19:34Z 19.1K followers, [--] engagements


"Anthropic's "Fast Mode" for Opus [---] is.*FAST*. Fixed a bug in [--] min. It can be used through your "extra usage" allowance. Was able to claim $50 in extra usage as a gift by going to the settings tab in the Claude app. Likely in preparation for this launch"  
[X Link](https://x.com/daniel_mac8/status/2020239795080032443)  2026-02-07T20:54Z 19.2K followers, 28.5K engagements


"Completing this one issue cost $1.68"  
[X Link](https://x.com/daniel_mac8/status/2020240439211880670)  2026-02-07T20:57Z 19.1K followers, [----] engagements


"@ibab Great for Apple's bottom line I guess"  
[X Link](https://x.com/daniel_mac8/status/2020241554758369301)  2026-02-07T21:01Z 19.1K followers, [----] engagements


"Anthropic is giving away super fast intelligence We granted all current Claude Pro and Max users $50 in free extra usage. This credit can be used on fast mode for Opus [---] in Claude Code. To use claim the credit and toggle on extra usage on https://t.co/C8CPCAvHYe. Then run claude update && claude and /fast. Enjoy We granted all current Claude Pro and Max users $50 in free extra usage. This credit can be used on fast mode for Opus [---] in Claude Code. To use claim the credit and toggle on extra usage on https://t.co/C8CPCAvHYe. Then run claude update && claude and /fast. Enjoy"  
[X Link](https://x.com/daniel_mac8/status/2020242616529674728)  2026-02-07T21:05Z 19.1K followers, [----] engagements


"@vlelyavin we'll see how their Super Bowl ad plays out. 😂"  
[X Link](https://x.com/daniel_mac8/status/2020254433771487484)  2026-02-07T21:52Z 19.1K followers, [---] engagements


"@natolambert are you not using auto-compaction"  
[X Link](https://x.com/daniel_mac8/status/2020282763471990890)  2026-02-07T23:45Z 19.1K followers, [---] engagements


"@diWRECKtion I'll put it on my landing page"  
[X Link](https://x.com/daniel_mac8/status/2020320848637575479)  2026-02-08T02:16Z 19.1K followers, [--] engagements


"@deepfates had to do a quick google to make sure this was a joke. phew"  
[X Link](https://x.com/daniel_mac8/status/2020323240892375426)  2026-02-08T02:26Z 19.1K followers, [---] engagements


"@cturnbull1968 Is that guy in the middle the one who made Mike Tyson throw a boxing match for money"  
[X Link](https://x.com/daniel_mac8/status/2020324747569975732)  2026-02-08T02:32Z 19.1K followers, [----] engagements


"@jakepaul you absolutely can call [---] or any other phone number regardless of how you feel about ICE and that is obvious"  
[X Link](https://x.com/daniel_mac8/status/2020328289710321928)  2026-02-08T02:46Z 19.2K followers, [----] engagements


"Opus [---] just gets me man"  
[X Link](https://x.com/daniel_mac8/status/2020331761515770033)  2026-02-08T03:00Z 19.2K followers, [----] engagements


"@modelsarereal depends on your definition but I personally define it as being able to continually update model weights without catastrophic forgetting"  
[X Link](https://x.com/daniel_mac8/status/2020502072752542092)  2026-02-08T14:16Z 19.1K followers, [---] engagements


"This is exactly what I mean by "Opus [---] just gets me." Was prepared to send the next message: "We need to make it human readable." Opus [---] beat me to it. Most "agentic" AI model yet. Opus [---] just gets me man. Opus [---] just gets me man"  
[X Link](https://x.com/daniel_mac8/status/2020506499022516631)  2026-02-08T14:34Z 19.2K followers, [----] engagements


"Opus [---] fast mode value from the perspective of an individual dev is the wrong way to think about it. Think about it from the scale of an enterprise. For example used to work at a CDP/Marketing Automation B2B SaaS co. If the data processing pipeline failed it could represent millions $ in lost revenue for our largest customers. So being able to fix an issue like that WAY FASTER can represent millions in $ value. It doesnt matter if it costs an extra $100 or even $1000 dollars at that scale. This is what Opus [---] fast mode is for. 2.5x faster 6x more expensive. Doesnt feel worth it imo."  
[X Link](https://x.com/daniel_mac8/status/2020525389060178262)  2026-02-08T15:49Z 19.2K followers, [----] engagements


"@AndrewCurran_ Will be interesting to see how strong the new models are though Im personally not too optimistic"  
[X Link](https://x.com/daniel_mac8/status/2020532686041854101)  2026-02-08T16:18Z 19.2K followers, [---] engagements


"@RapidResponse47 @sunofwisdom Doubtful hes a real Loser if hes on the USA Olympic ski team. Which should be obvious"  
[X Link](https://x.com/daniel_mac8/status/2020534032455303270)  2026-02-08T16:23Z 19.2K followers, [---] engagements


"@WallStreetMav She said nothing about the U.S.A as a country. She said it's been tough under this administration. If the administration doesn't want to get trashed they should stop being trash. Also all of MAGA should stop whining about it"  
[X Link](https://x.com/daniel_mac8/status/2020548851040440591)  2026-02-08T17:22Z 19.2K followers, [---] engagements


"Google prepares Gemini [--] GA launch. Consider it to be Gemini [---]. Should be on par with Opus [---] + GPT-5.2. Won't be long now"  
[X Link](https://x.com/daniel_mac8/status/2020558246298214906)  2026-02-08T18:00Z 19.3K followers, 91.6K engagements


"@chatgpt21 When are we getting [---] Chris"  
[X Link](https://x.com/daniel_mac8/status/2020578623116214773)  2026-02-08T19:20Z 19.2K followers, [----] engagements


"@chatgpt21 Oh today Super Bowl surprise"  
[X Link](https://x.com/daniel_mac8/status/2020584517850046624)  2026-02-08T19:44Z 19.2K followers, [---] engagements


"False alarm. @daniel_mac8 Hi from the team This flag has nothing to do with model's preview-to-GA move. It was just an internal flag we used while making preview version available to more customers. Now it's available by default unless your enterprise disables preview Sorry to spoil the speculation. 😂 @daniel_mac8 Hi from the team This flag has nothing to do with model's preview-to-GA move. It was just an internal flag we used while making preview version available to more customers. Now it's available by default unless your enterprise disables preview Sorry to spoil the speculation. 😂"  
[X Link](https://x.com/daniel_mac8/status/2020602088397656423)  2026-02-08T20:54Z 19.2K followers, [----] engagements


"Is this legal Introducing Claude with Ads Free access to Anthropics most powerful model powered by ads. Try it now: https://t.co/QU43cgIJRm Introducing Claude with Ads Free access to Anthropics most powerful model powered by ads. Try it now: https://t.co/QU43cgIJRm"  
[X Link](https://x.com/daniel_mac8/status/2020665594106740746)  2026-02-09T01:06Z 19.2K followers, 73.8K engagements


"@micsolana Music is the language of emotion Mike"  
[X Link](https://x.com/daniel_mac8/status/2020676762946937158)  2026-02-09T01:50Z 19.2K followers, [----] engagements


"@DrewPavlou Literally just a national humiliation ritual for the United States designed to make white people feel bad for existing. Youre talking about the Kid Rock halftime show right"  
[X Link](https://x.com/daniel_mac8/status/2020868203115679930)  2026-02-09T14:31Z 19.2K followers, [----] engagements


"Are you ready for GPT-5.3 aka Garlic 🧄 Mark Chen went on Ashlee Vance's "Core Memory" end of last year and said OpenAI researcher's "figured some things out" wrt pretraining. Expect those improvements with GPT-5.3. Plus we know the GPT-5.3-Codex was trained on Nvidia Blackwells. The first model from a major provider trained on the latest Nvidia chips. That makes a difference. Also my prediction is that it will have a very high GDPVal win rate and you'll be able to use the Codex desktop app to do all kinds of knowledge work outside coding. This new model is a big deal. Time to get hyped."  
[X Link](https://x.com/daniel_mac8/status/2020892283898306957)  2026-02-09T16:07Z 19.3K followers, 36.7K engagements


"@itslirrato @Polymarket Yikes. Im sure a strike is coming. Not sure if it will be today. Hopefully its limited"  
[X Link](https://x.com/daniel_mac8/status/2020935715669024831)  2026-02-09T18:59Z 19.2K followers, [---] engagements


"OpenAI's Super Bowl ad didn't promote ChatGPT. It promoted Codex. Speaks volumes about where the value will come from. More than [--] million people downloaded Codex App in the first week. 60+% growth in overall Codex user last week We'll keep Codex available to Free/Go users after this promotion; we may have to reduce limits there but we want everyone to be able to try Codex and start building. More than [--] million people downloaded Codex App in the first week. 60+% growth in overall Codex user last week We'll keep Codex available to Free/Go users after this promotion; we may have to reduce"  
[X Link](https://x.com/daniel_mac8/status/2020983192925372899)  2026-02-09T22:08Z 19.3K followers, 11.2K engagements


"GPT-5.3-Codex is the cracked engineer whos kind of a dick. You give him a task and then get outta there. Opus [---] is good but also a chill guy. Matters less how excellent the code is because youre just having fun working with him. This is what Ive learned after a few days"  
[X Link](https://x.com/daniel_mac8/status/2021007460765696463)  2026-02-09T23:45Z 19.2K followers, 16.2K engagements


"@macintoch haha love that analogy"  
[X Link](https://x.com/daniel_mac8/status/2021017038433767925)  2026-02-10T00:23Z 19.2K followers, [---] engagements


"@deredleritt3r So basically every support role besides actual lawyers"  
[X Link](https://x.com/daniel_mac8/status/2021027488626016274)  2026-02-10T01:04Z 19.2K followers, [----] engagements


"He really didn't like the idea of going to the Moon before Mars did he I resigned from xAI today. This company - and the family we became - will stay with me forever. I will deeply miss the people the warrooms and all those battles we have fought together. It's time for my next chapter. It is an era with full possibilities: a small team armed I resigned from xAI today. This company - and the family we became - will stay with me forever. I will deeply miss the people the warrooms and all those battles we have fought together. It's time for my next chapter. It is an era with full possibilities:"  
[X Link](https://x.com/daniel_mac8/status/2021196772056629466)  2026-02-10T12:17Z 19.2K followers, [----] engagements


"This is what Gary Marcus sees. Deep learning is hitting a wall https://t.co/atOJ2Rl8hb Deep learning is hitting a wall https://t.co/atOJ2Rl8hb"  
[X Link](https://x.com/daniel_mac8/status/2021232463020687606)  2026-02-10T14:39Z 19.2K followers, 24.3K engagements


"@thsottiaux The model is amazing the UX could use some love. For example one thing I love about Claude Code is that it predicts your next message and is often quite accurate in what I plan to do next. Things like that would improve Codex UX/DX"  
[X Link](https://x.com/daniel_mac8/status/2021233519440105912)  2026-02-10T14:43Z 19.2K followers, [---] engagements


"@obsdmd Hook that baby up to Codex/Claude Code and let her rip"  
[X Link](https://x.com/daniel_mac8/status/2021254834838511749)  2026-02-10T16:08Z 19.3K followers, [----] engagements


"@E__Strobel @obsdmd The main advantage is that you can now do it programatically. For example if you are working in another project using CC and have context in Obsidian that is relevant you can instruct CC to get the context using the Obsidian CLI. AI Agents are very good at using CLIs"  
[X Link](https://x.com/daniel_mac8/status/2021260472272482679)  2026-02-10T16:30Z 19.3K followers, [---] engagements


"@Fedesco5 haha I have no idea I muted him a long time ago. it's just funny to me"  
[X Link](https://x.com/daniel_mac8/status/2021260642728939985)  2026-02-10T16:31Z 19.2K followers, [---] engagements


"Assuming this prediction is correct (should find out this week) its clarifies OpenAIs unique strategy: GPT-5.3-Codex for business GPT-5o for pleasure OpenAI is the only lab differentiating on the model level to serve different users. It could pay off. My ultra reliable sources say GPT-5.3 is a red herring. Something different is coming this week from OpenAI. CNBC did report it's a "chat model". PREDICTION: GPT-5o with native multi-modal capabilities and a brand new voice input/output. My ultra reliable sources say GPT-5.3 is a red herring. Something different is coming this week from OpenAI."  
[X Link](https://x.com/daniel_mac8/status/2021265591047262691)  2026-02-10T16:50Z 19.3K followers, 29K engagements


"@the_real_jdizzy Very myopic view. At least in Yanns case hes still trying to do research and make discoveries"  
[X Link](https://x.com/daniel_mac8/status/2021282234402054195)  2026-02-10T17:56Z 19.2K followers, [---] engagements


"Spoke with a well-known AI startup last week about a Forward Deployed Engineer role. Hiring Manager said they have [--] full-time FDEs. Want to expand to [--] by end of July. *60* by the end of the year. Andrew is right that AI is going to *MASSIVELY* disrupt the knowledge work labor market. You best hedge is to develop AI skills. It doesn't need to be in a tech setting either. Could be: Legal Construction Medicine Any existing business vertical. Be the expert who genuinely knows how to generate value by applying AI in those settings. That is where you'll find all the new opportunity. Its"  
[X Link](https://x.com/daniel_mac8/status/2021310598907310452)  2026-02-10T19:49Z 19.3K followers, 20.2K engagements


"@InfraScaler It's definitely in that direction. @grok explain what a Forward Deployed Engineer does"  
[X Link](https://x.com/daniel_mac8/status/2021321382265573679)  2026-02-10T20:32Z 19.2K followers, [---] engagements


"@InfiniteHexx For me that would fall under the "business" category. Codex is not only for coding"  
[X Link](https://x.com/daniel_mac8/status/2021329824401264701)  2026-02-10T21:05Z 19.2K followers, [---] engagements


"GPT-5.3-Codex is inadequately named. Everybody knows it. It's the same scenario as Claude Code before Cowork. 5.3-Codex is NOT only for coding. It's great for any type of knowledge work that benefits from logical-reasoning. Esp in the Codex app with the right skills. @daniel_mac8 5.3-Codex is just meant for coding not knowledge work right @daniel_mac8 5.3-Codex is just meant for coding not knowledge work right"  
[X Link](https://x.com/daniel_mac8/status/2021349336429494724)  2026-02-10T22:23Z 19.3K followers, 19.6K engagements


"@jontweetssports Reeses Blizzards do rule though"  
[X Link](https://x.com/daniel_mac8/status/2021387689652523011)  2026-02-11T00:55Z 19.3K followers, [----] engagements


"When both kids are napping and I finally have some time to get some housework done"  
[X Link](https://x.com/daniel_mac8/status/2021390986681987259)  2026-02-11T01:09Z 19.3K followers, [----] engagements


"@Yuchenj_UW Is it called xAI because its mostly ex-employees"  
[X Link](https://x.com/daniel_mac8/status/2021393482355753283)  2026-02-11T01:18Z 19.3K followers, [----] engagements


"Now we know: The "x" in xAI stands for ex-employee"  
[X Link](https://x.com/daniel_mac8/status/2021413661366243721)  2026-02-11T02:39Z 19.3K followers, [----] engagements


"just got fired from xai. was the hr business partner responsible for making sure people dont leave xai"  
[X Link](https://x.com/daniel_mac8/status/2021562834228838687)  2026-02-11T12:31Z 19.3K followers, [----] engagements


"@gailcweiner Be here all day 😂"  
[X Link](https://x.com/daniel_mac8/status/2021564512915316909)  2026-02-11T12:38Z 19.3K followers, [---] engagements


"@Art_If_Ficial Word on the street is that they may have vested a good amount of equity which got converted to SpaceX equity during the merger so basically became very wealthy overnight"  
[X Link](https://x.com/daniel_mac8/status/2021600941032509594)  2026-02-11T15:03Z 19.3K followers, [---] engagements


"@Art_If_Ficial lol some day our ship will come in"  
[X Link](https://x.com/daniel_mac8/status/2021602015327989913)  2026-02-11T15:07Z 19.3K followers, [---] engagements


"@sama GPT-5.3-Codex is insanely good and the Codex app is a great user experience"  
[X Link](https://x.com/daniel_mac8/status/2021607490090147897)  2026-02-11T15:29Z 19.3K followers, [----] engagements


"This is a very interesting essay. Saw an English prof. criticizing it. Probably jealous because he's never written anything that's been seen by 35M people. I'm sure Matt used AI to help him write it. That's the spirit of the essay: AI can already help you do real valuable work. It's getting better at that FAST. That criticism is exactly the same as the people who got mad at Bob Dylan when he picked up the electric guitar. Anyway my takeaway is this: The Singularity is here. We are in the slow takeoff. Prepare yourself but not from a place of fear and anxiety. From a place of accepting reality"  
[X Link](https://x.com/daniel_mac8/status/2021615257123164660)  2026-02-11T16:00Z 19.3K followers, [----] engagements


"@nabeelqu Its probably less malice and more incompetence. They didnt underhand Covid and they dont understand AI"  
[X Link](https://x.com/daniel_mac8/status/2021626839832334691)  2026-02-11T16:46Z 19.3K followers, [---] engagements


"@mckaywrigley the singularity is here"  
[X Link](https://x.com/daniel_mac8/status/2021691165729333548)  2026-02-11T21:01Z 19.3K followers, [----] engagements


"Impossible to exaggerate the effectiveness of: Opus [---] + Claude Code joyously and quickly generating code. GPT-5.3-Codex coldly and calculatingly reviewing your code. The dynamic duo. Good cop. Bad cop"  
[X Link](https://x.com/daniel_mac8/status/2021705828042940547)  2026-02-11T22:00Z 19.3K followers, [----] engagements


"@NeilMcDevitt_ With great power comes great responsibility Neil 😂"  
[X Link](https://x.com/daniel_mac8/status/2021784214501245025)  2026-02-12T03:11Z 19.3K followers, [---] engagements


"@r0ck3t23 Damn this fool just jacked my post. Like my post instead. I wrote it with my own brain cells instead of AI. Elon: By the end of [----] AI will be able to directly generate binary. Bypassing the need to generate code. Hook that model up to a Neuralink. You've got imagination-2-software. Things will get weird. Not necessarily bad. But weird. https://t.co/VXKtMVMD9g Elon: By the end of [----] AI will be able to directly generate binary. Bypassing the need to generate code. Hook that model up to a Neuralink. You've got imagination-2-software. Things will get weird. Not necessarily bad. But"  
[X Link](https://x.com/daniel_mac8/status/2021904956220117267)  2026-02-12T11:11Z 19.3K followers, [---] engagements


"@ScienceEdgeHQ @r0ck3t23 It is the clip that I edited myself with the exact same captions and uses the exact same analogies that I used but embellished with AI generated writing. Stolen"  
[X Link](https://x.com/daniel_mac8/status/2021909199438135551)  2026-02-12T11:28Z 19.2K followers, [--] engagements


"@PhiloGroves I mean you would be able to test for correctness same way you do with compiled/interpreted code"  
[X Link](https://x.com/daniel_mac8/status/2021911901840838832)  2026-02-12T11:38Z 19.3K followers, [--] engagements


"I don't think there's anything inherently wrong with using AI to write. It's not a moral failing. But I do believe you are doing yourself a disservice in the long-term. The day is near when AI will be so much better than any human at generating automated interesting content. It will be oversupplied soon if not already. When that happens the only rare and valuable resource left from a creator perspective will be the ability to originate and express your genuine human thoughts and feelings. And if you constantly use AI to simulate that you'll lose the ability to do it. Trung Phan discovered the"  
[X Link](https://x.com/daniel_mac8/status/2021920965337215230)  2026-02-12T12:14Z 19.3K followers, [----] engagements


"@AdamHoltererer lol why"  
[X Link](https://x.com/daniel_mac8/status/2021924568567615836)  2026-02-12T12:29Z 19.3K followers, [--] engagements


"@AdamHoltererer This all makes sense but I doubt youll be required to generate binary directly just something youll eventually be able to do if you want"  
[X Link](https://x.com/daniel_mac8/status/2021925423907758371)  2026-02-12T12:32Z 19.3K followers, [--] engagements


"Big things are happening. Tibo who works on Codex at OpenAI says they are going to launch something that brings joy AND profoundly effects how you get things done. Soon Could be today Some things we make are purely meant to bring joy others will profoundly affect how you get things done. We are getting ready to launch something that combines both and the team is so excited to share it soon. ✨ Some things we make are purely meant to bring joy others will profoundly affect how you get things done. We are getting ready to launch something that combines both and the team is so excited to share it"  
[X Link](https://x.com/daniel_mac8/status/2021965137494659517)  2026-02-12T15:10Z 19.3K followers, [----] engagements


"@Makuh90 That would be amazing"  
[X Link](https://x.com/daniel_mac8/status/2021965710227075274)  2026-02-12T15:12Z 19.3K followers, [---] engagements


"@chatgpt21 I'll take it. Going skiing this weekend. Might do a little vibing on the slopes"  
[X Link](https://x.com/daniel_mac8/status/2021971657175248974)  2026-02-12T15:36Z 19.3K followers, [----] engagements


"@Conor_D_Dart huge music appreciator here"  
[X Link](https://x.com/daniel_mac8/status/2021974348203200862)  2026-02-12T15:47Z 19.2K followers, [--] engagements


"@OneStoryRoad let's go"  
[X Link](https://x.com/daniel_mac8/status/2021977982089515453)  2026-02-12T16:01Z 19.3K followers, [--] engagements


"🔥 Gemini [--] Deep Think got an upgrade and it is now likely the most intelligent AI model available. 84.6% and SOTA on ARC-AGI-2 Prediction: A Gemini model will be used to find the cure to Cancer. Weve upgraded our specialized reasoning mode Gemini [--] Deep Think to help solve modern science research and engineering challenges pushing the frontier of intelligence. 🧠 Watch how the Wang Lab at Duke University is using it to design new semiconductor materials. 🧵 https://t.co/BgSEmv00JP Weve upgraded our specialized reasoning mode Gemini [--] Deep Think to help solve modern science research and"  
[X Link](https://x.com/daniel_mac8/status/2021987887462723772)  2026-02-12T16:40Z 19.3K followers, [----] engagements


"✨ OpenAI releases GPT-5.3-Codex-Spark. It is the first iteration on OpenAI's partnership with ultra-fast inference provider Cerebras offering speeds of 1000/tps. It does have a lower SWE-Bench Pro score than GPT-5.3-Codex: 51.5% for Spark vs. 56.8% regular Codex on xhigh reasoning Use it for things you need done quick that are not too complex. GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj"  
[X Link](https://x.com/daniel_mac8/status/2022012415781617729)  2026-02-12T18:18Z 19.3K followers, [----] engagements


"Claude Sonnet [--] incoming. Are you ready for Opus [---] level coding abilities at Sonnet prices Get ready. Big week for Anthropic fans coming up😉 (Or perhaps just anyone who uses AI to code) Big week for Anthropic fans coming up😉 (Or perhaps just anyone who uses AI to code)"  
[X Link](https://x.com/daniel_mac8/status/2017656706692501878)  2026-01-31T17:50Z 19.3K followers, 124.5K engagements


"Claude Sonnet [--] released this week by Anthropic. 82.1% SWE-Bench $3/1m input + $15/1m output (same as Sonnet 4.5) MUCH faster than Opus [---] I hate to say this.but it will be *wild*. Esp in Claude Code. That's my prediction. Rumors say Sonnet [--] will be better than Opus [---] for sure not just as good Rumors say Sonnet [--] will be better than Opus [---] for sure not just as good"  
[X Link](https://x.com/daniel_mac8/status/2017950270739648949)  2026-02-01T13:16Z 19.3K followers, 461.8K engagements


"listen Im not the type of guy to write a sonnet [--] hype post [--] days in a row and be wrong but if I were Id say: sonnet [--] is coming today. it is by far the most intelligent ai model sonnet [--] drops tomorrow and i've heard from three separate sources inside anthropic that the benchmarks they're sitting on would mass-retire every model released in [----]. they delayed it twice because the safety team couldn't explain why it started solving problems it wasn't sonnet [--] drops tomorrow and i've heard from three separate sources inside anthropic that the benchmarks they're sitting on would mass-retire"  
[X Link](https://x.com/daniel_mac8/status/2019386027111293398)  2026-02-05T12:22Z 19.3K followers, 22.7K engagements


"Could have simply said: "No AGI until Continual Learning." Which some say arrives this year. AGI is definitely coming. AGI is not coming. We are nowhere near AGI. What we have today is inference not learning. Models get trained once on huge fixed datasets then frozen. You ask questions they remix patterns they already saw. Nothing updates. Nothing sticks. Talking to the model does not make AGI is not coming. We are nowhere near AGI. What we have today is inference not learning. Models get trained once on huge fixed datasets then frozen. You ask questions they remix patterns they already saw."  
[X Link](https://x.com/daniel_mac8/status/2020484268749639919)  2026-02-08T13:06Z 19.3K followers, 34.1K engagements


"My ultra reliable sources say GPT-5.3 is a red herring. Something different is coming this week from OpenAI. CNBC did report it's a "chat model". PREDICTION: GPT-5o with native multi-modal capabilities and a brand new voice input/output. BREAKING 🚨: GPT-5.3 mention has been spotted in the Codex PR. Considering that GitHub and Warp are preparing for big updates tomorrow we may see it very soon already. https://t.co/TtKqsaO53t BREAKING 🚨: GPT-5.3 mention has been spotted in the Codex PR. Considering that GitHub and Warp are preparing for big updates tomorrow we may see it very soon already."  
[X Link](https://x.com/daniel_mac8/status/2021049118114840890)  2026-02-10T02:30Z 19.3K followers, 119.8K engagements


"@tunguz It's weird that working from scratch often beats working with existing code. Bullish for newer startups vs. legacy cos"  
[X Link](https://x.com/daniel_mac8/status/2021981365194174722)  2026-02-12T16:14Z 19.3K followers, [---] engagements


"@shill_collin @OpenAI I believe it's also a smaller model to begin with. Doesn't score as high on coding benchmarks but much faster"  
[X Link](https://x.com/daniel_mac8/status/2022026103477666285)  2026-02-12T19:12Z 19.3K followers, [---] engagements


"Can you believe we used to do all this thinking with our own brain"  
[X Link](https://x.com/daniel_mac8/status/2022094954802426337)  2026-02-12T23:46Z 19.3K followers, [----] engagements


"@NeilMcDevitt_ it's so hard to do"  
[X Link](https://x.com/daniel_mac8/status/2022095954913243399)  2026-02-12T23:50Z 19.3K followers, [--] engagements


"@Makuh90 Sure have"  
[X Link](https://x.com/daniel_mac8/status/2022105765000229126)  2026-02-13T00:29Z 19.3K followers, [--] engagements


"@Vtrivedy10 Yes Its a hosted implementation of the ACE paper"  
[X Link](https://x.com/daniel_mac8/status/2022136267631423600)  2026-02-13T02:30Z 19.3K followers, [---] engagements


"@ToddKuehnl Yes exactly A hosted app of the Stanford paper"  
[X Link](https://x.com/daniel_mac8/status/2022157225670906235)  2026-02-13T03:53Z 19.3K followers, [---] engagements


"@xogegrok 🫡"  
[X Link](https://x.com/daniel_mac8/status/2022161643439059424)  2026-02-13T04:11Z 19.3K followers, [--] engagements


"@SaidAitmbarek Its a web app/MCP server you can connect to your agent. Go to you can check out the docs. Official launch coming up in a couple hours. http://aceagent.io http://aceagent.io"  
[X Link](https://x.com/daniel_mac8/status/2022288031676289252)  2026-02-13T12:33Z 19.3K followers, [--] engagements


"@macintoch I will fix it. Its because I have it set for first time customers only but then it doesnt work if you started with the trial"  
[X Link](https://x.com/daniel_mac8/status/2022479631941542009)  2026-02-14T01:14Z 19.3K followers, [--] engagements


"@macintoch Amazing share any feedback that you want"  
[X Link](https://x.com/daniel_mac8/status/2022479707380621378)  2026-02-14T01:15Z 19.3K followers, [--] engagements


"@dedene Oh yes Claude lovvves discombobulating you can actually customize those messages now"  
[X Link](https://x.com/daniel_mac8/status/2022747442375200987)  2026-02-14T18:59Z 19.3K followers, [--] engagements


"Elon: By the end of [----] AI will be able to directly generate binary. Bypassing the need to generate code. Hook that model up to a Neuralink. You've got imagination-2-software. Things will get weird. Not necessarily bad. But weird"  
[X Link](https://x.com/daniel_mac8/status/2021783057645416592)  2026-02-12T03:06Z 19.3K followers, 15.7K engagements


"@ToddKuehnl Im using GPT-5.2 Thinking for all the agents right now. May adjust at some point"  
[X Link](https://x.com/daniel_mac8/status/2022350158835445764)  2026-02-13T16:40Z 19.3K followers, [--] engagements


"https://x.com/daniel_mac8/status/2022365907662864647s=20 🔥 INTRODUCING: ACE The Agentic Context Engineer ACE is an AI Agent that improves context for any AI Agent. Context gets better every time you use ACE. No fine-tuning. No manual prompt editing. Just use ACE and your AI Agents get smarter. 🧵 👇 https://t.co/4NdMe5lhmt https://x.com/daniel_mac8/status/2022365907662864647s=20 🔥 INTRODUCING: ACE The Agentic Context Engineer ACE is an AI Agent that improves context for any AI Agent. Context gets better every time you use ACE. No fine-tuning. No manual prompt editing. Just use ACE and your"  
[X Link](https://x.com/daniel_mac8/status/2022371090983428394)  2026-02-13T18:03Z 19.3K followers, [---] engagements


"@DanielGPT2022 Thats cool. You could import these into ACE and improve them"  
[X Link](https://x.com/daniel_mac8/status/2022676934531125373)  2026-02-14T14:18Z 19.3K followers, [--] engagements


"@Mollyploofkins Youre not allowed to do that"  
[X Link](https://x.com/daniel_mac8/status/2022711954020958679)  2026-02-14T16:38Z 19.3K followers, [----] engagements


"🔥 🤖 🕵♂ INTRODUCING: Meta Agent 💥 this is by far the most exciting thing I've created using AI made using Manus and @windsurf_ai 🔑 Meta Agent is an AI agent that uses the OpenAI Agents SDK to GENERATE NEW AGENTS you simply tell Meta Agent in natural language the agent you want. 🤯 and Meta Agent generates it for you fully open -- link to GitHub and example walkthrough below"  
[X Link](https://x.com/daniel_mac8/status/1900735621036937709)  2025-03-15T02:27Z 19.3K followers, 150K engagements


"@disclosetv say it aint so claude say it aint so"  
[X Link](https://x.com/daniel_mac8/status/2022462198715683126)  2026-02-14T00:05Z 19.3K followers, [----] engagements


"Moral/ethical qualms aside an AIs ability to assist in military operations is a very good real-world benchmark. The Pentagon is considering cutting ties with Anthropic after the company refused to allow its AI models to be used for all lawful purposes insisting on strict bans around mass domestic surveillance and fully autonomous weapons. Tensions escalated following a disputed episode https://t.co/ZfNVPKYvQ2 The Pentagon is considering cutting ties with Anthropic after the company refused to allow its AI models to be used for all lawful purposes insisting on strict bans around mass domestic"  
[X Link](https://x.com/daniel_mac8/status/2023052357911867425)  2026-02-15T15:10Z 19.3K followers, [----] engagements


"Claude Code with Claude for Chrome is a dev superpower. Allows CC to open up a browser tab and validate frontend features itself. Best tip Boris has shared yet. 10s video on how to configure it. 13/ A final tip: probably the most important thing to get great results out of Claude Code -- give Claude a way to verify its work. If Claude has that feedback loop it will 2-3x the quality of the final result. Claude tests every single change I land to https://t.co/XJ8WxOxjo0 13/ A final tip: probably the most important thing to get great results out of Claude Code -- give Claude a way to verify its"  
[X Link](https://x.com/daniel_mac8/status/2009977701772472642)  2026-01-10T13:16Z 19.3K followers, 173K engagements


"@joon_s_pk @karpathy @drfeifei @adamdangelo @rauchg @scottbelsky Will Simile prove that we are already in a simulation"  
[X Link](https://x.com/daniel_mac8/status/2022044137860870469)  2026-02-12T20:24Z 19.3K followers, [---] engagements


"Is Anthropic cooked GPT-5.1-Codex tops Sonnet [---] on SWE-Bench at 26x cheaper. Gemini [--] incoming. Don't know how good it will be at code. Likely amazing. Opus [---] *needs* to hit hard at a great price. Anthropic has trouble with cost efficiency. (No joy in saying this)"  
[X Link](https://x.com/daniel_mac8/status/1989719345002430804)  2025-11-15T15:37Z 19.3K followers, 122.6K engagements


"Something Big is Happening. Tomorrow. ACE - Agentic Context Engineer. Create playbooks Observe outcomes Improve playbooks through evolution Improve any AI Agent. For any task. The more you use it. Soon"  
[X Link](https://x.com/daniel_mac8/status/2022111703757103417)  2026-02-13T00:52Z 19.3K followers, 14.5K engagements


"ACE was built entirely with a combo of Claude Code + Codex. Didnt write a single line of code. Anyone saying you cant create complex software with AI is absolutely full of it. Exhibit A: http://aceagent.io 🔥 INTRODUCING: ACE The Agentic Context Engineer ACE is an AI Agent that improves context for any AI Agent. Context gets better every time you use ACE. No fine-tuning. No manual prompt editing. Just use ACE and your AI Agents get smarter. 🧵 👇 https://t.co/4NdMe5lhmt http://aceagent.io 🔥 INTRODUCING: ACE The Agentic Context Engineer ACE is an AI Agent that improves context for any AI"  
[X Link](https://x.com/daniel_mac8/status/2022449023056253263)  2026-02-13T23:13Z 19.3K followers, [----] engagements


"@DrBoneLove Thanks Dr. BoneLove ❤"  
[X Link](https://x.com/daniel_mac8/status/2022465582445334923)  2026-02-14T00:19Z 19.3K followers, [--] engagements


"How to build a $1mil ARR software business [--]. Claude Code + Codex [--]. Lie about it on the internet to make yourself seem more successful than you are ACE did get [--] real customers on launch day. ACE autonomously improves your AI Agents context. Try: http://aceagent.io 🔥 INTRODUCING: ACE The Agentic Context Engineer ACE is an AI Agent that improves context for any AI Agent. Context gets better every time you use ACE. No fine-tuning. No manual prompt editing. Just use ACE and your AI Agents get smarter. 🧵 👇 https://t.co/4NdMe5lhmt http://aceagent.io 🔥 INTRODUCING: ACE The Agentic Context"  
[X Link](https://x.com/daniel_mac8/status/2022663621772480820)  2026-02-14T13:26Z 19.3K followers, [----] engagements


"@GuntherEagleman In which part specifically is Amanda calling for Trump to be censored here"  
[X Link](https://x.com/daniel_mac8/status/2022696343496786245)  2026-02-14T15:36Z 19.3K followers, [---] engagements


"@bartek_wl Yes it did. And Im not selling anything"  
[X Link](https://x.com/daniel_mac8/status/2022684805700464808)  2026-02-14T14:50Z 19.3K followers, [--] engagements


"this is an amazing way to think about prompting o1 from @benhylak"  
[X Link](https://x.com/anyuser/status/1878283032215408886)  2025-01-12T03:29Z 19.3K followers, 2.1M engagements


"🚨 Did Moonshot just crack Continual Learning Moonshot introduced: 'Checkpoint Engine'. It rapidly updates model weights in LLM inference engines. Can update a 1T param model like Kimi-K2 in [--] seconds. If this works it truly is a *game changer* BIG. IF. TRUE. 🤯"  
[X Link](https://x.com/anyuser/status/1965789965859930491)  2025-09-10T14:50Z 19.3K followers, 113.2K engagements


"Gemini [--] is coming. This we know. What we don't know is how good it will be. My prediction is that it will be clearly the best AI model available both vibes and benchmark based. Google has the momentum now and I don't think anyone is stopping that train now it's a rollin'"  
[X Link](https://x.com/anyuser/status/1971019345334780007)  2025-09-25T01:10Z 19.3K followers, 44.6K engagements


"This is laughable. BBC produced a study with extremely negative findings for AI Assistants. Look at what models they used: GPT-4o () Gemini [---] Flash () Preplexity - Free Tier () MS Copilot - Free Tier () This is nothing more than Baby-Boomer negativity biased amygdala hacking attention harvesting. Its completely self-serving. Actually I take it back its not laughable. Its just sad. https://www.bbc.co.uk/mediacentre/2025/new-ebu-research-ai-assistants-news-content https://www.bbc.co.uk/mediacentre/2025/new-ebu-research-ai-assistants-news-content"  
[X Link](https://x.com/anyuser/status/1981708377605496926)  2025-10-24T13:04Z 19.3K followers, 127.5K engagements


"2026: Sholto Douglas Anthropic predicts the Claude Code experience but for *ALL* forms of knowledge work. Few realize that Claude Code isn't only about code Few realize that Claude Code isn't only about code"  
[X Link](https://x.com/daniel_mac8/status/2004643379120849022)  2025-12-26T20:00Z 19.3K followers, 106.4K engagements


"GPT-5.3-Codex and Opus [---]. Two frontier coding models in one day. Opus [---] clearly excels in some areas. GPT-5.3-Codex in others. Will use both extensively. P.S. it *is* comical how each lab avoids reporting benchmarks it knows the other is strong in"  
[X Link](https://x.com/daniel_mac8/status/2019477876169404583)  2026-02-05T18:27Z 19.3K followers, 64.3K engagements


"SaaS is *not* dead. Agree with Jensen. But the incumbents like ServiceNow and SAP They will get smoked. AI will bring: More software More people using software More people consuming software And it will be absolutely nothing like old SaaS. It will be *alive*. https://twitter.com/i/web/status/2020320216652148786 https://twitter.com/i/web/status/2020320216652148786"  
[X Link](https://x.com/daniel_mac8/status/2020320216652148786)  2026-02-08T02:14Z 19.3K followers, [----] engagements


"🎯 ACE fixes this with three agents working in a loop: [--]. Generator provides a playbook to your agent [--]. Reflector analyzes what worked and what didn't [--]. Curator makes surgical edits that preserve what's already good remove what's not Your playbook gets better. Automatically. From real outcomes. https://twitter.com/i/web/status/2022365911047360691 https://twitter.com/i/web/status/2022365911047360691"  
[X Link](https://x.com/daniel_mac8/status/2022365911047360691)  2026-02-13T17:43Z 19.3K followers, [---] engagements


"🔑 The workflow is dead simple: [--]. Create a playbook (your AI instructions/context) [--]. Connect any agent via MCP (Claude Code Codex custom agents whatever supports MCP) [--]. Agent reads playbook does task records outcome [--]. ACE evolves the playbook based on real results That's it. The loop runs itself. https://twitter.com/i/web/status/2022365912901333088 https://twitter.com/i/web/status/2022365912901333088"  
[X Link](https://x.com/daniel_mac8/status/2022365912901333088)  2026-02-13T17:43Z 19.3K followers, [---] engagements


"@NeilMcDevitt_ Instant karma"  
[X Link](https://x.com/daniel_mac8/status/2022413839548231919)  2026-02-13T20:53Z 19.3K followers, [---] engagements


"Here's the problem ACE solves: AI Agents work better when given the right context and right instructions. But how often do you update your CLAUDE .md/AGENTS .md With ACE it happens automatically. Autonomously editing and improving the context available to your AI Agent"  
[X Link](https://x.com/daniel_mac8/status/2022365909650702677)  2026-02-13T17:43Z 19.3K followers, [---] engagements


"📊 From the research this is built on: +10.6% improvement on agent tasks 87% lower latency vs existing methods 75-91% fewer tokens And it compounds. Every evolution makes the next one better"  
[X Link](https://x.com/daniel_mac8/status/2022365914121769274)  2026-02-13T17:43Z 19.3K followers, [---] engagements


"Starts at $9/mo. Free trial available. First [---] Founding Members get 20% off for life when subscribing to the Pro/Ultra plan with code: FOUNDING20 If you're building with AI Agents ACE is how your instructions should work. http://aceagent.io http://aceagent.io"  
[X Link](https://x.com/daniel_mac8/status/2022365915296199044)  2026-02-13T17:43Z 19.3K followers, [---] engagements


"This is amazing. Jeff and team created a way for AI agents to discover ever better memory mechanisms for themselves. Definitely a form of Recursive Self-Improvement. One of the coolest results I've seen in a while. 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"  
[X Link](https://x.com/daniel_mac8/status/2021684295056728354)  2026-02-11T20:34Z 19.3K followers, 44.2K engagements


"@karpathy Are you just vibing your way to the Singularity Andrej"  
[X Link](https://x.com/daniel_mac8/status/2021694812689211661)  2026-02-11T21:16Z 19.3K followers, [----] engagements


"@OpenAI GPT-5.3-Codex-Spark is powered by Cerebras with speeds of 1000/tps for those who are curious:"  
[X Link](https://x.com/daniel_mac8/status/2022010923247108389)  2026-02-12T18:12Z 19.3K followers, 17.1K engagements


"GPT-5.3-Codex-Spark is insane. ⚡ Intelligence velocity is the #1 underrated facet of intelligence. Even if one is slightly less smart if you can think 1000x faster you win. This is a coding model. Extrapolate this to the automated AI researcher that OpenAI is building. Super fast superintelligent and always on. That is when the fast takeoff starts. That is coming. Watch this [--] min video of GPT-5.3-Codex-Spark in action with a workflow tip. https://twitter.com/i/web/status/2022052736930336830 https://twitter.com/i/web/status/2022052736930336830"  
[X Link](https://x.com/daniel_mac8/status/2022052736930336830)  2026-02-12T20:58Z 19.3K followers, 44.7K engagements


"@elonmusk @AnthropicAI Bro what are you talking about I'm a heterosexual white male and me and Claude are boys. Claude gets me"  
[X Link](https://x.com/daniel_mac8/status/2022061360998449305)  2026-02-12T21:32Z 19.3K followers, 13.1K engagements


"I'm sorry. GPT-5.3-Codex-Spark might be trash. The code it writes is terrible. GPT-5.3-Codex had to correct everything. More testing is needed. Standby. It's likely that you should only use it to rapidly generate code after you've used a smarter model to create a structure. GPT-5.3-Codex-Spark is insane. ⚡ Intelligence velocity is the #1 underrated facet of intelligence. Even if one is slightly less smart if you can think 1000x faster you win. This is a coding model. Extrapolate this to the automated AI researcher that OpenAI is building. https://t.co/mE2cgZ0kED GPT-5.3-Codex-Spark is insane."  
[X Link](https://x.com/daniel_mac8/status/2022161078197829682)  2026-02-13T04:09Z 19.3K followers, 37.8K engagements


"🔥 INTRODUCING: ACE The Agentic Context Engineer ACE is an AI Agent that improves context for any AI Agent. Context gets better every time you use ACE. No fine-tuning. No manual prompt editing. Just use ACE and your AI Agents get smarter. 🧵 👇"  
[X Link](https://x.com/daniel_mac8/status/2022365907662864647)  2026-02-13T17:43Z 19.3K followers, 19.9K engagements


"Dario Amodei told Dwarkesh: "I'm at 90% we get to a country of geniuses in a datacenter within [--] years." The only reason he won't go higher is because the world is so unpredictable. China could invade Taiwan or all the datacenters could get bombed he says. SOMETHING BIG IS HAPPENING. Notice: the people trying to convince you otherwise make their living as "AI experts" who've been harvesting your attention telling you it's a nothingburger while the exponential continues to march on. https://twitter.com/i/web/status/2022415402173689891 https://twitter.com/i/web/status/2022415402173689891"  
[X Link](https://x.com/daniel_mac8/status/2022415402173689891)  2026-02-13T20:59Z 19.3K followers, 15.9K engagements


"@chatgpt21 @dwarkesh_sp Sholto Douglas said late last year: "Continual Learning solved in a satisfactory way in 2026." Likely talking about the former that Dario mentions here"  
[X Link](https://x.com/daniel_mac8/status/2022436567361675755)  2026-02-13T22:23Z 19.3K followers, [----] engagements


"Hegseth: Claude create a plan to capture Nicolas Maduro with no casualties. Clear the context and auto-accept edits. Claude: *Whatchamacalliting* Breaking: The Pentagon used Anthropics AI tool Claude in its military operation to capture former Venezuelan President Nicols Maduro https://t.co/HwXRnafUo0 Breaking: The Pentagon used Anthropics AI tool Claude in its military operation to capture former Venezuelan President Nicols Maduro https://t.co/HwXRnafUo0"  
[X Link](https://x.com/daniel_mac8/status/2022482048544625123)  2026-02-14T01:24Z 19.3K followers, [----] engagements


"Dario Amodei: "We are trying to keep a 10x revenue curve going." "There is zero time for bullshit." "There is zero time for feeling like we're productive when we're not." "These tools make us more productive." Soon this productivity will hit the entire economy"  
[X Link](https://x.com/daniel_mac8/status/2022503920430682601)  2026-02-14T02:51Z 19.3K followers, 14.7K engagements


"This is such a big deal. GPT-5.2 discovered new scientific knowledge. This is way more important than landing on the moon. Every important innovation over the last century follows from discoveries in Physics. GPT-5.2 derived a new result in theoretical physics. Were releasing the result in a preprint with researchers from @the_IAS @VanderbiltU @Cambridge_Uni and @Harvard. It shows that a gluon interaction many physicists expected would not occur can arise under specific GPT-5.2 derived a new result in theoretical physics. Were releasing the result in a preprint with researchers from @the_IAS"  
[X Link](https://x.com/daniel_mac8/status/2022647734361337918)  2026-02-14T12:22Z 19.3K followers, [----] engagements


"@rockorager Where do you think the value is Good ideas"  
[X Link](https://x.com/daniel_mac8/status/2022658332373455343)  2026-02-14T13:05Z 19.3K followers, [---] engagements


"our internal model 👀 Very excited about the "First Proof" challenge. I believe novel frontier research is perhaps the most important way to evaluate capabilities of the next generation of AI models. We have run our internal model with limited human supervision on the ten proposed problems. The Very excited about the "First Proof" challenge. I believe novel frontier research is perhaps the most important way to evaluate capabilities of the next generation of AI models. We have run our internal model with limited human supervision on the ten proposed problems. The"  
[X Link](https://x.com/daniel_mac8/status/2022693323992158368)  2026-02-14T15:24Z 19.3K followers, [----] engagements


"They said the same thing when Prometheus stole fire from the gods. These are obviously not earth-shattering results but the ability to produce genuinely new knowledge however small is a significant milestone and I hope we all take it seriously with excitement and caution. These are obviously not earth-shattering results but the ability to produce genuinely new knowledge however small is a significant milestone and I hope we all take it seriously with excitement and caution"  
[X Link](https://x.com/daniel_mac8/status/2022793421484687733)  2026-02-14T22:01Z 19.3K followers, [----] engagements


"@polynoamial @GoogleDeepMind Inference compute is all you need"  
[X Link](https://x.com/daniel_mac8/status/2022819289535820197)  2026-02-14T23:44Z 19.3K followers, [---] engagements


"@Trav_Tayl0r @disclosetv Woulda said Claude would never do that but now Im not sure"  
[X Link](https://x.com/daniel_mac8/status/2022824940886806779)  2026-02-15T00:07Z 19.3K followers, [--] engagements


"@elbouzzz So its a good test for ASI"  
[X Link](https://x.com/daniel_mac8/status/2023126454473728090)  2026-02-15T20:05Z 19.3K followers, [--] engagements


"@elonmusk @StephenKing Its a fair point. Hate to be that guy but both sides need to cut out the demonization of the other end of the spectrum. A functioning society needs a balance of conservative and liberal forces. Right now its like the left hand trying to cut off the right"  
[X Link](https://x.com/daniel_mac8/status/2023128924738527474)  2026-02-15T20:14Z 19.3K followers, [----] engagements


"there is zero chance these DeepSeek-V4 Thinking benchmarks are real And there is zero chance Chinese models are out in front of US models. im sorry. there is zero chance. guys. the new deepseek model is much further in front out of any US model than i expected. theyll be strictly locking down access china only. theres absolutely no chance we ever catch back up. unless were hiding models that are 10x more capable. its over. not sure guys. the new deepseek model is much further in front out of any US model than i expected. theyll be strictly locking down access china only. theres absolutely no"  
[X Link](https://x.com/daniel_mac8/status/2023134504022581638)  2026-02-15T20:37Z 19.3K followers, [----] engagements

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