# ![@raw_works Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::2229107533.png) @raw_works Raymond Weitekamp

Raymond Weitekamp posts on X about claude code, pi, if you, o1 the most. They currently have [-----] followers and [---] posts still getting attention that total [---] engagements in the last [--] hours.

### Engagements: [---] [#](/creator/twitter::2229107533/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2229107533/c:line/m:interactions.svg)

- [--] Week [-------] +24,939%
- [--] Month [-------] +9,641%
- [--] Months [-------] +806%
- [--] Year [-------] +880%

### Mentions: [--] [#](/creator/twitter::2229107533/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2229107533/c:line/m:posts_active.svg)

- [--] Months [--] no change
- [--] Year [--] +40%

### Followers: [-----] [#](/creator/twitter::2229107533/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2229107533/c:line/m:followers.svg)

- [--] Week [-----] +35%
- [--] Month [-----] +40%
- [--] Months [-----] +74%
- [--] Year [-----] +157%

### CreatorRank: [---------] [#](/creator/twitter::2229107533/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2229107533/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[technology brands](/list/technology-brands)  [stocks](/list/stocks)  [finance](/list/finance)  [cryptocurrencies](/list/cryptocurrencies)  [social networks](/list/social-networks)  [celebrities](/list/celebrities)  [currencies](/list/currencies)  [countries](/list/countries)  [vc firms](/list/vc-firms)  [gaming](/list/gaming) 

**Social topic influence**
[claude code](/topic/claude-code), [pi](/topic/pi) #1419, [if you](/topic/if-you), [o1](/topic/o1), [money](/topic/money), [llm](/topic/llm), [cloudflare](/topic/cloudflare), [code](/topic/code), [end of](/topic/end-of), [ai](/topic/ai)

**Top assets mentioned**
[Cloudflare, Inc. (NET)](/topic/cloudflare) [Alphabet Inc Class A (GOOGL)](/topic/$googl) [Propbase (PROPS)](/topic/$props) [FilesCoins Power Cu (FILECOIN)](/topic/files) [CENTS (CENTS)](/topic/cents) [DeepSeek (DEEPSEEK)](/topic/deepseek) [Runes Coin (RUNES)](/topic/runes) [Fluid (FLUID)](/topic/fluid) [Microsoft Corp. (MSFT)](/topic/microsoft)
### Top Social Posts
Top posts by engagements in the last [--] hours

"@jordibruin Whisper large isnt very good. The main thing about HappyScribe was the custom vocabulary - which is really helpful for proper nouns such as names of companies/products/people. With MacWhisper I have to fix this or prompt a LLM to fix this"  
[X Link](https://x.com/raw_works/status/1733654179397398823)  2023-12-10T01:05Z [---] followers, [--] engagements


"Experimenting with using my social media accounts as 'repeaters':"  
[X Link](https://x.com/raw_works/status/1736602691365675266)  2023-12-18T04:22Z [---] followers, [--] engagements


"There's nothing like extremely low blood sugar and a COVID infection to get your priorities straight"  
[X Link](https://x.com/raw_works/status/1736965079516504503)  2023-12-19T04:22Z [---] followers, [---] engagements


"@disiok time for $100M series a"  
[X Link](https://x.com/raw_works/status/1757526426075701589)  2024-02-13T22:05Z [---] followers, [--] engagements


"@nomic_ai @LangChainAI @llama_index @MongoDB 🦙🦙🦙"  
[X Link](https://x.com/raw_works/status/1757810352371233171)  2024-02-14T16:53Z [---] followers, [---] engagements


"@nomic_ai @qdrant_engine @MongoDB @epsilla_inc @WeAviate @langchain4j @DataStax too plz (astradb)"  
[X Link](https://x.com/raw_works/status/1757810672576938281)  2024-02-14T16:54Z [---] followers, [--] engagements


"@antonosika i'm hungry to see someone run grok-1 for evalplus:"  
[X Link](https://x.com/raw_works/status/1771260992510460090)  2024-03-22T19:41Z [---] followers, [--] engagements


"I am very excited to bring the #texttoCAD power of neThingXYZ to @poe_platform Now you access the @PolySpectra #textto3D #genai directly from our MakeItReal bot on Poe. Make it real. https://poe.com/MakeItReal https://poe.com/MakeItReal"  
[X Link](https://x.com/raw_works/status/1781419536031646114)  2024-04-19T20:28Z [---] followers, [--] engagements


"so excited to see this expand globally my first hackathon ever was at the OG AGI House with @poe_platform in Nov. .and now i've officially drunk the koolaid. Launch-a-thon stream with @generalcatalyst and @rajivayyangar https://t.co/I4j91VFdA3 Launch-a-thon stream with @generalcatalyst and @rajivayyangar https://t.co/I4j91VFdA3"  
[X Link](https://x.com/raw_works/status/1797093447637103070)  2024-06-02T02:30Z [---] followers, [---] engagements


"@aditya_advani [--] blue [--] brown"  
[X Link](https://x.com/raw_works/status/1797119630433218905)  2024-06-02T04:14Z [---] followers, [--] engagements


"@alexalbert__ @sumbhavsethia would love to see the "sandbox" backend for how this was implemented"  
[X Link](https://x.com/raw_works/status/1804237017787830302)  2024-06-21T19:36Z [---] followers, [--] engagements


""moatless" is not only a leading AI coding agent by @aorwall it is also a truly deep metaphor for our swiftly evolving civilization. "moatless" is the evolution of the cathedral and the bazaar. what do you do when you have no moat you dance"  
[X Link](https://x.com/raw_works/status/1805418462363501029)  2024-06-25T01:51Z [---] followers, [--] engagements


"@rauchg @vercel @v0 @nextjs Make v0 write Svelte(kit) instead of react. Cc @SvelteSociety"  
[X Link](https://x.com/raw_works/status/1842225123098501392)  2024-10-04T15:27Z [---] followers, [--] engagements


"Everyone is obsessed with NotebookLM & the new "Canvas" from OpenAI. Yes they are cool but in my opinion @cursor_ai is actually a much more powerful tool for co-writing with AI. In this video I show how I get from a poor quality transcription of a voice memo and a brief markdown outline to a draft of a blog post I'm doing for @WevolverApp - all with one prompt and in a few seconds. I've been using Cursor "off-label" as a co-writer of articles grant proposals and other highly technical documents for about a year now and it has been an insane productivity booster"  
[X Link](https://x.com/raw_works/status/1842237696959267118)  2024-10-04T16:17Z [---] followers, [----] engagements


"amazing Now can someone sveltify this @SvelteSociety @sveltejs @Rich_Harris @huntabyte Introducing sidebar.tsx25 components to help you build all kinds of sidebars. I don't like building sidebars. So I built 30+ of them. All types. Then simplified the core into sidebar.tsxa strong foundation to build on top of. It works with Next.js Remix Vite & Laravel. https://t.co/wfVqT1aMVY Introducing sidebar.tsx25 components to help you build all kinds of sidebars. I don't like building sidebars. So I built 30+ of them. All types. Then simplified the core into sidebar.tsxa strong foundation to build on"  
[X Link](https://x.com/raw_works/status/1847714983464903061)  2024-10-19T19:02Z [---] followers, [----] engagements


"w00t I love shadcn & svelte as much as I hate web development. (Which is a lot) shadcn-svelte Sidebars ✅ Just need to add the docs and update the registry/cli and we'll be clear for takeoff 🚣♂ shadcn-svelte Sidebars ✅ Just need to add the docs and update the registry/cli and we'll be clear for takeoff 🚣♂"  
[X Link](https://x.com/raw_works/status/1848458301240909839)  2024-10-21T20:16Z [---] followers, [--] engagements


"all day today: Cannot use export let in runes mode use $props() instead i don't get it. @AnthropicAI had [--] whole days to train sonnet [---] to speak @sveltejs [--] i think we deserve another update to sonnet for this momentous occasion. cc @SvelteSociety @cursor_ai"  
[X Link](https://x.com/raw_works/status/1849879811927216219)  2024-10-25T18:24Z [---] followers, [--] engagements


"@richardfzhang @connorjewiss Honestly X has such a horrible API and seems to penalize tweets made by APII think an X API would be popular"  
[X Link](https://x.com/raw_works/status/1851410085659984317)  2024-10-29T23:45Z [---] followers, [--] engagements


"@rauchg @v0 Svelte [--] apps duh cc: @Rich_Harris"  
[X Link](https://x.com/raw_works/status/1857557391597777282)  2024-11-15T22:52Z [---] followers, [--] engagements


"cursor tip that is especially relevant for new frameworks like svelte 5: [--]. add the new docs as custom doc [--]. tell the AI to generate the key things to remember about this framework. [--]. then tell it to write those to .cursorrules voila. AI teaching AI. (learned the hard way after constantly reminding sonnet to re-read the @sveltejs docs)"  
[X Link](https://x.com/raw_works/status/1859314019460358261)  2024-11-20T19:13Z [---] followers, [---] engagements


"@mckaywrigley Funny that I found out from your tweet and not from [--] months of being on the waitlist"  
[X Link](https://x.com/raw_works/status/1866681083879886991)  2024-12-11T03:07Z [---] followers, [--] engagements


"@FireworksAI_HQ Does it allow for a prompt"  
[X Link](https://x.com/raw_works/status/1866682529052500066)  2024-12-11T03:12Z [---] followers, [--] engagements


"for the love of svelte could someone at @vercel teach v0 how to import shadcn-svelte"  
[X Link](https://x.com/raw_works/status/1868837976983359649)  2024-12-17T01:57Z [---] followers, [--] engagements


"@kmiddleton14 @vercel @v0 @aisdk congrats can you teach @v0 svelte 5"  
[X Link](https://x.com/raw_works/status/1869041063383683330)  2024-12-17T15:24Z [---] followers, [---] engagements


"is o1 smart enough to CAD see for yourself day one support for o1 in neThingXYZ"  
[X Link](https://x.com/raw_works/status/1869390485082763436)  2024-12-18T14:33Z [---] followers, [--] engagements


"🤯the new "specdec" llama [---] 70B from @GroqInc can edit CAD files faster than i can think"  
[X Link](https://x.com/raw_works/status/1869447308196339903)  2024-12-18T18:19Z [---] followers, [----] engagements


"None of the above. Please ask your consortium how to make this compatible w OpenAI client chat completions. Maybe needs a local proxy server that becomes baseURL and then the consortiums name becomes model name. (Then we can use consortiums in any app that allows baseURL to be changed.)"  
[X Link](https://x.com/raw_works/status/1870901932455588326)  2024-12-22T18:39Z [---] followers, [--] engagements


"@allhands_ai from my experience with cursor o1 has better ideas but tends not to follow the instructions for output format. so i'm often using o1 to come up with the plan and then switching to sonnet to implement the code. also check this related effort out = https://aider.chat/2024/12/21/polyglot.html https://aider.chat/2024/12/21/polyglot.html"  
[X Link](https://x.com/raw_works/status/1876069238831771912)  2025-01-06T00:52Z [---] followers, [---] engagements


"@ivanfioravanti Are you noticing a lot of failures / timeouts Both direct and on openrouter over the last week Ive been having reliability issues where sometimes it just doesnt respond at all. Others on cline and aider discords noticing the same thing"  
[X Link](https://x.com/raw_works/status/1876277395734532366)  2025-01-06T14:39Z [---] followers, [--] engagements


"on the @thursdai_pod @joaomdmoura shared that o1 has a really hard time with function calling (tool use) that it tries to make up the answer without calling the tools. it doesn't surprise me that it struggles with a tool-heavy harness like @allhands_ai but is good w aider @aidan_mclau newsonnet is really good at understanding that it's in a harness. incredible work by anthropic. It does 50+ steps just fine in our system having a set of tools + many invocations to solve a problem is far from o1's training regime where it tries to reason to the final answer @aidan_mclau newsonnet is really good"  
[X Link](https://x.com/raw_works/status/1876431404864885047)  2025-01-07T00:51Z [---] followers, [--] engagements


"@edgarrmcgregor Whats the cross streets (My parents still in Altadena near Eaton canyon trying to help find more info.)"  
[X Link](https://x.com/raw_works/status/1877052882471514336)  2025-01-08T18:00Z [---] followers, [---] engagements


"@kestrel4547 @GavinNewsom @realDonaldTrump @elonmusk What street is that Cant read the sign"  
[X Link](https://x.com/raw_works/status/1877055636074074453)  2025-01-08T18:11Z [---] followers, [--] engagements


"welp like a lot of my friends the house i grew up in is totally gone. my heart goes out to beautiful altadena"  
[X Link](https://x.com/raw_works/status/1877091394994528542)  2025-01-08T20:33Z [---] followers, [---] engagements


"@vikhyatk wow congrats hey i haven't had much luck with "handwriting ocr" on moondream in the past is that an area you are interested in any tips (i realize there are a lot of different use cases to try to cover for vlms)"  
[X Link](https://x.com/raw_works/status/1877416892287090774)  2025-01-09T18:07Z [---] followers, [---] engagements


"@CerebrasSystems i have my credit card ready for you where is the pay-per-token plan 😂 currently only groq will take my money"  
[X Link](https://x.com/raw_works/status/1877418223278133346)  2025-01-09T18:12Z [---] followers, [----] engagements


"i think the @AnthropicAI 'memory' MCP needs a little bit of work lol. 🤣 did anyone test this"  
[X Link](https://x.com/raw_works/status/1878818470873456655)  2025-01-13T14:56Z [---] followers, [---] engagements


"if you are using o1 or o1 pro at all.definitely read the article linked below to get your money's worth. o1 is mind-blowing when you know how to use it. it's really not a chat model -- you have to think of it more like a "report generator" (link to article below) https://t.co/LQCyUDHGm4 o1 is mind-blowing when you know how to use it. it's really not a chat model -- you have to think of it more like a "report generator" (link to article below) https://t.co/LQCyUDHGm4"  
[X Link](https://x.com/raw_works/status/1878879827916370051)  2025-01-13T19:00Z [---] followers, [---] engagements


"@jonas so the llm just needs to decide whether to bif or bof at any given step"  
[X Link](https://x.com/raw_works/status/1879940588734382503)  2025-01-16T17:15Z [---] followers, [---] engagements


"@SullyOmarr Aider copy/paste mode is great too for giving o1 pro repo context. Added benefit that you can go back to using aider w other models (including o1) afterwards"  
[X Link](https://x.com/raw_works/status/1880051869499973746)  2025-01-17T00:37Z [---] followers, [---] engagements


"@xundecidability @Hailuo_AI @OpenRouterAI @simonw it is real on the direct minimax api but unfortunately @OpenRouterAI is having issues with timeouts when you get close to the 1M"  
[X Link](https://x.com/raw_works/status/1880694502459056332)  2025-01-18T19:11Z [---] followers, [---] engagements


"@aorwall All for a nickel right"  
[X Link](https://x.com/raw_works/status/1881447683820552583)  2025-01-20T21:04Z [---] followers, [--] engagements


"@aorwall Or $1800 for o1 ;)"  
[X Link](https://x.com/raw_works/status/1881449798693470458)  2025-01-20T21:12Z [---] followers, [--] engagements


"@DrJimFan this is karmic payback for openai taking o1 away from @openrouter (and still not shipping o1 api access to tier [--] after like [--] weeks)"  
[X Link](https://x.com/raw_works/status/1881512738545095079)  2025-01-21T01:22Z [---] followers, [----] engagements


"I couldn't find a good arena or leaderboard for image to 3D AIs. .so I just made my own. Inspired by LMSYS ImgSYS and text-to-image arenas the arena does head-to-head match-ups for image-to-3D genai. The leaderboard calculates the ELO score. Open to suggestions/submissions for both input images as well as image-to-3D models and products that should be included"  
[X Link](https://x.com/raw_works/status/1882412748912431394)  2025-01-23T12:59Z [---] followers, [---] engagements


"Arena: Leaderboard: Ironically the open source model trellis (which is also the cheapest to run only a few cents) is currently in the lead.but with very low N.so please cast your votes https://nething.xyz/leaderboard https://nething.xyz/arena https://nething.xyz/leaderboard https://nething.xyz/arena"  
[X Link](https://x.com/raw_works/status/1882412751580287261)  2025-01-23T12:59Z [---] followers, [--] engagements


"@ozenhati Like since before deepseek was cool I swear check your support ticket logs 🤪"  
[X Link](https://x.com/raw_works/status/1882576285983228173)  2025-01-23T23:48Z [---] followers, [--] engagements


"@theo depends on what time today o3-mini launches"  
[X Link](https://x.com/raw_works/status/1885357114425594109)  2025-01-31T15:58Z [---] followers, [---] engagements


"@vikhyatk I just want an LLM that can write with Svelte [--] runes syntax @SvelteSociety"  
[X Link](https://x.com/raw_works/status/1888009984883007961)  2025-02-07T23:40Z [---] followers, [---] engagements


"@nahkhsinad @sidahuj Cool Keep me posted Stoked on your CAD idea Joined the discord. I built as a text-to-CAD tool. The biggest weakness is that no one wants to switch their CAD tool so I think this idea of bringing AI to the apps people already use is the way foward. http://neThing.xyz http://neThing.xyz"  
[X Link](https://x.com/raw_works/status/1901657492880245064)  2025-03-17T15:30Z [---] followers, [--] engagements


"it's alive we have liftoff for multiplayer positions. vibe coding night [--] for @levelsio fun challenge after cleaning up baby barf - yay. between running a company by day being a father to a 1-yo and [--] game dev experience.this will have to fly on vibes alone"  
[X Link](https://x.com/raw_works/status/1901825882441850883)  2025-03-18T02:39Z [---] followers, [---] engagements


"major failure of the night: wasting 90min and $18 on colyseus cloud thinking that the multiplayer positions was going to be a severe headache.but unfortunately ai doesn't know how to use it. major win: sonnet [---] on cursor is really f-ing good at using cloudflare wrangler. refactored to durable objects in a few minutes"  
[X Link](https://x.com/raw_works/status/1901827254222811183)  2025-03-18T02:45Z [---] followers, [--] engagements


"@aisdk and maybe svelte [--] support too ;) @sveltejs"  
[X Link](https://x.com/raw_works/status/1903118238554128525)  2025-03-21T16:15Z [---] followers, [---] engagements


"not quite ready to take down autodesk but this is pretty cool Introducing SynCity 🌆 SynCity generates entire 3D worlds from a text prompt with no training or optimisation. It leverages pretrained 2D and 3D generators and generates scenes on a grid tile by tile. The generated 3D environments are diverse fully coherent and navigable. 🧵👇 https://t.co/64W8C3atOs Introducing SynCity 🌆 SynCity generates entire 3D worlds from a text prompt with no training or optimisation. It leverages pretrained 2D and 3D generators and generates scenes on a grid tile by tile. The generated 3D environments are"  
[X Link](https://x.com/raw_works/status/1903192267860386106)  2025-03-21T21:09Z [---] followers, [---] engagements


"@nicolaiklemke @FAL you say "just FAL wrappers" as if that's a bad thing. such amazing company to be in. @poe_platform is just a FAL wrapper soon @Adobe will be just a FAL wrapper i for one am proud to be a FAL wrapper"  
[X Link](https://x.com/raw_works/status/1904270537527840943)  2025-03-24T20:34Z [---] followers, [---] engagements


"@DreamStarter_1 super cool concept rip expensive rendering software"  
[X Link](https://x.com/raw_works/status/1904986138265485636)  2025-03-26T19:57Z [---] followers, [----] engagements


"@levelsio now that we have vibe coding i don't understand why everyone doesn't just switch to hugo. english is the new gui"  
[X Link](https://x.com/raw_works/status/1907897220222902552)  2025-04-03T20:45Z [---] followers, [---] engagements


"This is currently the best explanation of a Saturday release for a major model. My theory was simply that Zuck forgot to watch the ending of Ready Player One where they shut down their metaverse for two days a week to let people make out with each other in real life. Its very common for leadership at top labs to be in the know of other labs release schedules so the simplest explanation to Meta releasing today is that next week is going to be bonkers or at least plausibly outshine llama [--] (and they wanted to release last week but other Its very common for leadership at top labs to be in the"  
[X Link](https://x.com/raw_works/status/1908649737550512487)  2025-04-05T22:35Z [---] followers, [---] engagements


"@levelsio definitely don't go to china if you fail to find the h100 you seek here's a build writeup for 4090s/5090s from @a16z https://a16z.com/building-an-efficient-gpu-server-with-nvidia-geforce-rtx-4090s-5090s/ https://a16z.com/building-an-efficient-gpu-server-with-nvidia-geforce-rtx-4090s-5090s/"  
[X Link](https://x.com/raw_works/status/1910773401267978292)  2025-04-11T19:14Z [---] followers, [---] engagements


"@grok [--]. there is nothing in the ui with the words "studio".so this is very confusing. i was able to get bad flappy bird game but the UI has no indication this is possible. [--]. how do i connect google drive on the web app i see no buttons for this. http://grok.com http://grok.com"  
[X Link](https://x.com/raw_works/status/1912503896376856736)  2025-04-16T13:50Z [---] followers, [--] engagements


"@willccbb r2 baby where you at @deepseek_ai"  
[X Link](https://x.com/raw_works/status/1921978190320337385)  2025-05-12T17:18Z [---] followers, [---] engagements


"anyone need optimizable llms.txt dspy incepted me into incepting omar into doing this @raw_works @DSPyOSS @damekdavis @RepoPrompt In defense of @raw_works s I agents I wrote that program: @raw_works @DSPyOSS @damekdavis @RepoPrompt In defense of @raw_works s I agents I wrote that program:"  
[X Link](https://x.com/raw_works/status/1922443992739836346)  2025-05-14T00:08Z [---] followers, [---] engagements


"for those who simply cannot let go of their [----] prompting skillz i learned a pro @DSPyOSS tip from @cyrusnewday: put your favorite prompt in the description the way i think of this you are "incepting" the ai program with your prompt but letting the program re-word the actual prompt. @DSPyOSS but you see the problem with dspy is that now i will actually have to know what i want the ai to do. .you took away all of the thumb-twiddling. @DSPyOSS but you see the problem with dspy is that now i will actually have to know what i want the ai to do. .you took away all of the thumb-twiddling"  
[X Link](https://x.com/raw_works/status/1924808192879452605)  2025-05-20T12:43Z [---] followers, [----] engagements


"@jxmnop so @rishi_d_jha and @jxmnop - would this be a more efficient means of distilling a smaller model from a larger teacher model for example: could i get o3 performance on tasks i care about in a more efficient way than current distillation methods"  
[X Link](https://x.com/raw_works/status/1925567236762124602)  2025-05-22T14:59Z [---] followers, [--] engagements


"i like this analogy not sure how true it is "Compared to the traditional ReAct method the interleaved thinking approach is faster and more fluid. You can think of ReAct like a detective taking notes and making deductions step by step whereas interleaved thinking is more like a domain expert mentally juggling facts while explaining a concept. In other words thought and action happen in one thinking block not another complete loop.""  
[X Link](https://x.com/raw_works/status/1934995568369012830)  2025-06-17T15:24Z [---] followers, [--] engagements


"@burkaygur @FAL woo"  
[X Link](https://x.com/raw_works/status/1940845830644421025)  2025-07-03T18:51Z [---] followers, [--] engagements


"hot fire at claude code camp rn from @every can't wait for the github repo"  
[X Link](https://x.com/raw_works/status/1946246920684368263)  2025-07-18T16:33Z [---] followers, [----] engagements


"@every @danshipper @bran_don_gell @DannyAziz97 @poojary_yash @alxai_ @kieranklaassen @naveennaidu_m @katelaurielee wow thanks for the vibetunnel recommendation @DannyAziz97 - so cool. great way to get outside (and would be cooler if iOS dictation engine didn't suck so much (or if apple let people override this for custom keyboards))"  
[X Link](https://x.com/raw_works/status/1946250913716305967)  2025-07-18T16:49Z [---] followers, [---] engagements


"@AmirAnvarov @every @danshipper @bran_don_gell @DannyAziz97 @poojary_yash @alxai_ @kieranklaassen @naveennaidu_m @katelaurielee i got an email by subscribing to @every newsletter. the session was recorded but i'm not sure if they posted the link yet. maybe @katelaurielee has the link"  
[X Link](https://x.com/raw_works/status/1947703040729076193)  2025-07-22T16:59Z [---] followers, [--] engagements


"do you track cost in this leaderboard the podcast notes made me think of this: "Wolfram did a benchmark run that would have cost him $150 using Claude Opus. On W&B Inference with the Qwen3-Thinking model it cost him [--] cents. That's not a typo. It's a game-changer for developers and researchers." @altryne"  
[X Link](https://x.com/raw_works/status/1951299768539287856)  2025-08-01T15:11Z [---] followers, [---] engagements


"@MaximeRivest to me this is the more useful speed benchmark because it averages many real users: exact values vary but @CerebrasSystems still smoking everyone else. it's so painful to go back to sonnet https://openrouter.ai/openai/gpt-oss-120b/providerssort=throughput https://openrouter.ai/openai/gpt-oss-120b/providerssort=throughput"  
[X Link](https://x.com/raw_works/status/1952838272255967388)  2025-08-05T21:05Z [---] followers, [---] engagements


"@emilecairncross @EastlondonDev @mntruell @CerebrasSystems groq doesn't serve qwen3 coder to my knowledge. and for the models where both serve them i like to sort openrouter by throughput: (yesterday cerebras was faster for this one today groq is faster 🤷♂) https://openrouter.ai/openai/gpt-oss-120b/providerssort=throughput https://openrouter.ai/openai/gpt-oss-120b/providerssort=throughput"  
[X Link](https://x.com/raw_works/status/1953176306473967768)  2025-08-06T19:28Z [---] followers, [---] engagements


"chatgpt dictation is amazing and simply refuses to spell "claude code" correctly cloud code clawd code clod code"  
[X Link](https://x.com/raw_works/status/1956032994914811974)  2025-08-14T16:39Z [---] followers, [--] engagements


"tired of banging my head against the wall with browser-use. claude code with the playwright mcp server nailed it on the first try. here's how to get web automation working: [--]. install: claude mcp add playwright -- npx -y @playwright/mcp@latest [--]. log into your target site manually in headed mode [--]. let claude read/interact with the page through structured browser APIs no more screenshot guessing games or flaky selectors. just works"  
[X Link](https://x.com/raw_works/status/1956036008249565472)  2025-08-14T16:51Z [---] followers, [---] engagements


"@LakshyAAAgrawal @DSPyOSS right now the only things i'm having trouble with are: (not ugly) wandb tracing and inference-time use case examples"  
[X Link](https://x.com/raw_works/status/1956163098408055142)  2025-08-15T01:16Z [---] followers, [---] engagements


"@swyx big model means big variance this is @emollick 's "jagged frontier""  
[X Link](https://x.com/raw_works/status/1957494245524177268)  2025-08-18T17:26Z [---] followers, [--] engagements


"@dosco what's your jules v claude code opinion"  
[X Link](https://x.com/raw_works/status/1957495088612839658)  2025-08-18T17:29Z [---] followers, [---] engagements


"@MatthewBerman i feel the same way.slowly trying to build something myself. chat isn't the right ux will let you know when i think it's worth $1000/mo ;)"  
[X Link](https://x.com/raw_works/status/1958683651719192919)  2025-08-22T00:12Z [---] followers, [--] engagements


"@corbtt @OpenPipeAI @CoreWeave clearly this must be part of the thursdai news tomorrow @WolframRvnwlf @thursdai_pod @altryne"  
[X Link](https://x.com/raw_works/status/1963405662316003686)  2025-09-04T00:56Z [---] followers, [--] engagements


"@rohanpaul_ai thanks for sharing looks dope especially if powered by super fast inference"  
[X Link](https://x.com/raw_works/status/1966313025520222426)  2025-09-12T01:28Z [---] followers, [---] engagements


"@ekzhang1 totally agree still using my PS3 everyday it is the center of my gratitude practice"  
[X Link](https://x.com/raw_works/status/1966320265693925520)  2025-09-12T01:57Z [---] followers, [---] engagements


"@dok2001 @Cloudflare agreed. many thanks to @dosco (Ax) so i can run DSPy on CF workers"  
[X Link](https://x.com/raw_works/status/1966520256383328433)  2025-09-12T15:12Z [---] followers, [---] engagements


"@willccbb lol doesn't really end at [--] years unfortunately. (still mooching on the family verizon plan)"  
[X Link](https://x.com/raw_works/status/1966606407752450274)  2025-09-12T20:54Z [---] followers, [---] engagements


"@MostafaRohani @OpenAI super impressive out of curiosity - what would this have cost at consumer token pricing"  
[X Link](https://x.com/raw_works/status/1968678221605384673)  2025-09-18T14:07Z [---] followers, [---] engagements


"@llm_wizard Lol yeah I felt like it never got started.like an hour of getting on the same page and then.cut"  
[X Link](https://x.com/raw_works/status/1972006833838768189)  2025-09-27T18:34Z [---] followers, [--] engagements


"@jespr @Shpigford Thank you I know nothing about rails but I want to learn I'm pretty comfortable with svelte and react so hopefully this is more friendly than jumping right into hotwire"  
[X Link](https://x.com/raw_works/status/1972434286528905508)  2025-09-28T22:52Z [---] followers, [--] engagements


"turns out the singularity was started by just putting claude code in a while loop. We ran Claude code in a while loop and it ported @browser_use to Typescript https://t.co/4g23oYN0zO We ran Claude code in a while loop and it ported @browser_use to Typescript https://t.co/4g23oYN0zO"  
[X Link](https://x.com/raw_works/status/1972486435891765664)  2025-09-29T02:19Z [---] followers, [---] engagements


"very creative cool to see anthropic bring gen ui directly to the people We're also releasing a temporary research preview called "Imagine with Claude". In this experiment Claude generates software on the fly. No functionality is predetermined; no code is prewritten. Available to Max users for [--] days. Try it out: https://t.co/z4soh1v4sJ We're also releasing a temporary research preview called "Imagine with Claude". In this experiment Claude generates software on the fly. No functionality is predetermined; no code is prewritten. Available to Max users for [--] days. Try it out:"  
[X Link](https://x.com/raw_works/status/1972719161572004079)  2025-09-29T17:44Z [---] followers, [---] engagements


"@nummanthinks @opencode @claudeai @thdxr anyone have instructions for configuring opencode with gemini vertex cli i have something i want to try where i need the rate limits that i have on vertex"  
[X Link](https://x.com/raw_works/status/1973380245274182062)  2025-10-01T13:31Z [---] followers, [--] engagements


"@osanseviero maybe gemini three can get vertex credentials working reliably in gemini-cli then i can start sending you all of my money as quickly as possible"  
[X Link](https://x.com/raw_works/status/1973467788695183585)  2025-10-01T19:19Z [---] followers, [---] engagements


"@thetobiaskrug @polar_sh @aisdk https://docs.stripe.com/agentsframework=vercel https://docs.stripe.com/agentsframework=vercel"  
[X Link](https://x.com/raw_works/status/1973471639854788882)  2025-10-01T19:34Z [---] followers, [--] engagements


"@JackWoth98 @googleaidevs This looks awesome One feature request - it's surprisingly hard to use the CLI with vertex credentials. I have a super high vertex rate limit and especially starting next week I would like to spend a lot of money on Gemini [--] w the CLI as fast as possible :)"  
[X Link](https://x.com/raw_works/status/1974165025020653641)  2025-10-03T17:29Z [---] followers, [---] engagements


"@JackWoth98 what is the best practice for doing headless with vertex credentials i keep getting logged out"  
[X Link](https://x.com/raw_works/status/1975554782741074315)  2025-10-07T13:32Z [---] followers, [--] engagements


"@MaximeRivest aquavoice can be configured to pull technical vocabulary and proper nouns (and even class/function names) from your active app. i don't think the dynamic custom vocab is that hard to do - especially not for a dspy wizard such as yourself"  
[X Link](https://x.com/raw_works/status/1975638822768779264)  2025-10-07T19:06Z [---] followers, [--] engagements


"@vikhyatk @cheatyyyy Bruh omarchy will install in [--] mins or your money back. @OmarchyLinux"  
[X Link](https://x.com/raw_works/status/1975970400246071638)  2025-10-08T17:03Z [---] followers, [--] engagements


"@vikhyatk i'm failing despite @simonw 's leak of the claude skills for microsoft word docs. maybe now i can get @browser_use to do one of those freetax filing websites for me. https://github.com/simonw/claude-skills/tree/main/mnt/skills/public https://github.com/simonw/claude-skills/tree/main/mnt/skills/public"  
[X Link](https://x.com/raw_works/status/1977461219473670234)  2025-10-12T19:47Z [---] followers, [--] engagements


"If you want to get better output telling your model You are an AI engineer specializing in production-grade LLM applications will not change that. 📢 Time for an update on my workflow. This one's a [--] min read so buckle up. 100% organic and hand-written like an animal. https://t.co/utYYfRM8vX 📢 Time for an update on my workflow. This one's a [--] min read so buckle up. 100% organic and hand-written like an animal. https://t.co/utYYfRM8vX"  
[X Link](https://x.com/raw_works/status/1978258071022424152)  2025-10-15T00:34Z [---] followers, [----] engagements


"Please dont tell me youre using Cursor Chat in your IDE the modern programming equivalent of competing with a pogo stick in a Formula One race. Watch yourself.that's Jensen's favorite pogo stick https://dataconomy.com/2025/10/15/jensen-huang-says-every-nvidia-engineer-now-codes-with-cursor/ https://dataconomy.com/2025/10/15/jensen-huang-says-every-nvidia-engineer-now-codes-with-cursor/ https://dataconomy.com/2025/10/15/jensen-huang-says-every-nvidia-engineer-now-codes-with-cursor/ https://dataconomy.com/2025/10/15/jensen-huang-says-every-nvidia-engineer-now-codes-with-cursor/"  
[X Link](https://x.com/raw_works/status/1978518967003316733)  2025-10-15T17:50Z [---] followers, [---] engagements


"@vikhyatk salesforce takes the specifications from the customers and brings them to the engineers"  
[X Link](https://x.com/raw_works/status/1978526707041681467)  2025-10-15T18:21Z [---] followers, [--] engagements


"@richardfzhang @InteguruAI any progress on one for x"  
[X Link](https://x.com/raw_works/status/1870219806890438747)  2024-12-20T21:28Z [----] followers, [--] engagements


"@karpathy great review but no api @xai what's a boy to do"  
[X Link](https://x.com/raw_works/status/1891965447852974298)  2025-02-18T21:38Z [---] followers, [--] engagements


"@nickwm even at [----] LOC the edit applies are really unreliable. is this still being worked on i know when i first met the cursor founders about a year ago fast edits was the top priority"  
[X Link](https://x.com/raw_works/status/1892535198685818969)  2025-02-20T11:22Z [---] followers, [--] engagements


"Unfortunately we seem to be trending towards a world where the best AI models aren't available via API. Fortunately Eric Provencher @pvncher made Repo Prompt. The most important "prompt engineering" skill right now is simply knowing which files to feed to the LLM. Repo Prompt makes that fun and easy (for both the humans and the LLMs). Hat tip to @paulgauthier for aider copy/paste mode which was my first exposure to this possibility as well as to @mckaywrigley for generously sharing his workflows"  
[X Link](https://x.com/raw_works/status/1892682908605022313)  2025-02-20T21:09Z [---] followers, [----] engagements


"@willccbb The other case where uv is the unambiguous winner is when youre setting up virtual machines for example on modal every time you change your dependency it has to rebuild the image and the difference between pip and uv pip is super stark"  
[X Link](https://x.com/raw_works/status/1893379503675646206)  2025-02-22T19:17Z [---] followers, [--] engagements


"does anyone have a workflow for giving cursor a video as input instructions i'm thinking: [--]. i make loom/zoom clips/etc screencast of using the app with voiceover instructions [--]. gemini (or some other video model) analyzes [--]. cursor implements the changes [--]. mezcal margaritas"  
[X Link](https://x.com/raw_works/status/1895077492571934848)  2025-02-27T11:44Z [----] followers, [---] engagements


"closest things i've found: - - cc @EastlondonDev https://github.com/eastlondoner/cursor-tools/issues/51 https://forum.cursor.com/t/feature-request-attaching-videos-and-multiple-images-to-gemini-models/13895 https://github.com/eastlondoner/cursor-tools/issues/51 https://forum.cursor.com/t/feature-request-attaching-videos-and-multiple-images-to-gemini-models/13895"  
[X Link](https://x.com/raw_works/status/1895077494505754806)  2025-02-27T11:44Z [----] followers, [--] engagements


"@andimarafioti any handwriting ocr in there"  
[X Link](https://x.com/raw_works/status/1964066709045694640)  2025-09-05T20:42Z [----] followers, [---] engagements


"first pinky we will use @DSPyOSS and @LiteLLM proxy server with custom guardrails to translate any api call into spanish. .then we try to take over the world"  
[X Link](https://x.com/raw_works/status/1972333693848211952)  2025-09-28T16:12Z [----] followers, [----] engagements


"@bernhardsson @modal @Lux_Capital @Redpoint @AmplifyPartners congrats lemme know if you need any help with your spend plan"  
[X Link](https://x.com/raw_works/status/1972682563903742035)  2025-09-29T15:19Z [---] followers, [---] engagements


"@MaximeRivest yes way better than puppeteer or playwright (also - you should checkout the deepwiki mcp can ask questions about any public gh repo)"  
[X Link](https://x.com/raw_works/status/1977476956099928528)  2025-10-12T20:50Z [---] followers, 14.3K engagements


"The launch today is kind of funny timing. I've been thinking a lot about how to use the new Chrome DevTools MCP server to do co-browsing with coding agents. I think there are probably a lot of use cases where that still might be smoother than Atlas. But if you could combine the two that would be really coolan agent talking to a browser that has an agent inside of it"  
[X Link](https://x.com/raw_works/status/1980740537566027986)  2025-10-21T20:58Z [---] followers, [--] engagements


"@rauchg nice +1 for svelte eval llms really suck at svelte"  
[X Link](https://x.com/raw_works/status/1981452333478236574)  2025-10-23T20:07Z [---] followers, [--] engagements


"@VikParuchuri which api endpoint is the best for handwriting OCR (i can't put images in the playground to test quickly)"  
[X Link](https://x.com/raw_works/status/1981453368309538894)  2025-10-23T20:11Z [---] followers, [--] engagements


"@Sikandar_Bhide @rauchg thanks for sharing your mcp looks useful. have you tried claude skills i think it would be cool to see svelte as a skill perhaps could get even better results than as a mcp"  
[X Link](https://x.com/raw_works/status/1983268101744680983)  2025-10-28T20:22Z [---] followers, [--] engagements


"@steipete cc my favorite agent peeps @AmpCode @FactoryAI @EnoReyes @ryancarson @thdxr @trq212 @dkundel - "time travel" seems like a very useful checkpointing feature"  
[X Link](https://x.com/raw_works/status/1984292658894745804)  2025-10-31T16:13Z [---] followers, [--] engagements


"@mitsuhiko @steipete @badlogicgames @Cloudflare most llms are really good at the wrangler cli"  
[X Link](https://x.com/raw_works/status/1985463363761996000)  2025-11-03T21:45Z [---] followers, [---] engagements


"@boristane @mitsuhiko @steipete @badlogicgames @Cloudflare you can config worker observability in wrangler https://chatgpt.com/share/690923d9-ddbc-8009-8adb-d6c3e8c09eea https://chatgpt.com/share/690923d9-ddbc-8009-8adb-d6c3e8c09eea"  
[X Link](https://x.com/raw_works/status/1985465166973051372)  2025-11-03T21:52Z [---] followers, [--] engagements


"here's my pitch: CLIs for DSPy use and optimize any coding agent with the power of DSPy. would appreciate any feedback on this idea from the @DSPyOSS peeps and especially @lateinteraction @MaximeRivest @a1zhang @LakshyAAAgrawal (links below)"  
[X Link](https://x.com/raw_works/status/1987179093746721285)  2025-11-08T15:23Z [---] followers, [----] engagements


"@pamelafox @simonw mostly trying to get it to test new tools for me"  
[X Link](https://x.com/raw_works/status/1987237811397836877)  2025-11-08T19:16Z [---] followers, [--] engagements


"@vikhyatk this might get me to finally leave os x. i just couldn't do it in omakub with the pasting in the terminal"  
[X Link](https://x.com/raw_works/status/1987280654187614621)  2025-11-08T22:06Z [---] followers, [--] engagements


"@vikhyatk hodl on those rsus champ"  
[X Link](https://x.com/raw_works/status/1987286348118773955)  2025-11-08T22:29Z [---] followers, [----] engagements


"for the multi-agent orchestration piece this is pretty cool: from @IndyDevDan (who doesn't seem to use x). the code is available if you buy his online course "agentic horizon" (which is confusingly connected to but separate from this one: ) https://agenticengineer.com/tactical-agentic-coding https://www.youtube.com/watchv=p0mrXfwAbCg https://www.youtube.com/watchv=p0mrXfwAbCg https://agenticengineer.com/tactical-agentic-coding https://www.youtube.com/watchv=p0mrXfwAbCg https://www.youtube.com/watchv=p0mrXfwAbCg"  
[X Link](https://x.com/raw_works/status/1987580236179923256)  2025-11-09T17:57Z [---] followers, [---] engagements


"@PaulSolt @steipete i've come to accept that ai news is impossible to keep up with. .but i would never have guessed that codex would make it impossible to keep up with a single ai-powered builder. don't stop @steipete (the singularity is nigh)"  
[X Link](https://x.com/raw_works/status/1987676663027474514)  2025-11-10T00:20Z [---] followers, [---] engagements


"@googledevs except why can't i use my vertex credentials (which have access) https://github.com/google-gemini/gemini-cli/issues/13284 https://github.com/google-gemini/gemini-cli/issues/13284"  
[X Link](https://x.com/raw_works/status/1990924359452487782)  2025-11-18T23:25Z [---] followers, [---] engagements


"what would you do with $100k in google cloud / vertex credits if you had to spend them in the next [--] months (asking for a friend)"  
[X Link](https://x.com/raw_works/status/1991209026881732859)  2025-11-19T18:16Z [----] followers, [---] engagements


"@altryne i guess this means agi is here"  
[X Link](https://x.com/raw_works/status/1993071795163824443)  2025-11-24T21:38Z [----] followers, [--] engagements


"@rsdgpt Someone give this guy $5M on a $50M pre money"  
[X Link](https://x.com/raw_works/status/1993128135651377531)  2025-11-25T01:22Z [----] followers, [--] engagements


"@FaroukAdeleke3 @aparnadhinak Yeah I don't know what this means either DSPy is open source and GEPA has a separate self contained repo. I think this is really interesting work but the "no DSPy lock-in" is not a good look"  
[X Link](https://x.com/raw_works/status/1993137762489569771)  2025-11-25T02:00Z [----] followers, [--] engagements


"@danshipper you are not alone. it is well documented by @Steve_Yegge and @RealGeneKim in their book: Vibe Coding i'm here for you"  
[X Link](https://x.com/raw_works/status/1993715838915142009)  2025-11-26T16:17Z [----] followers, [---] engagements


"@mattzcarey @interaction why tee them up when you can just drive your favorite coding cli entirely from whatsapp cc @steipete RIP @VibeTunnel https://github.com/steipete/warelay https://github.com/steipete/warelay"  
[X Link](https://x.com/raw_works/status/1993720098163671190)  2025-11-26T16:34Z [----] followers, [--] engagements


"@doodlestein i agree 100%. this is the "takeoff" moment. https://x.com/raw_works/status/1996610377522786434s=20 fast takeoff https://x.com/raw_works/status/1996610377522786434s=20 fast takeoff"  
[X Link](https://x.com/raw_works/status/1998441773086159347)  2025-12-09T17:17Z [----] followers, [--] engagements


"@dexhorthy @ChiragCX @jooshmayer somehow i can't get the dspy team to give a shit. if anyone wants to test out my fork it should work already. just swaps litellm for stdin = cli = stdout"  
[X Link](https://x.com/raw_works/status/1998913297287098764)  2025-12-11T00:30Z [----] followers, [--] engagements


"@steipete @clawdbot for the code you should try running with cerebras for light speed: https://x.com/stevekrouse/status/1990902186226684260s=20 ok but what if we generated a new app on every keystroke hope my friends at @cerebras don't hate me https://t.co/2NyzHm8vXY https://x.com/stevekrouse/status/1990902186226684260s=20 ok but what if we generated a new app on every keystroke hope my friends at @cerebras don't hate me https://t.co/2NyzHm8vXY"  
[X Link](https://x.com/raw_works/status/1999651145988321765)  2025-12-13T01:22Z [----] followers, [---] engagements


"@steipete @clawdbot if you want to keep the brains of opus but maybe speed up a bit these could maybe help.but inference is probably still the bottleneck: @morphllm @relace_ai both have 10k token/s code edit models"  
[X Link](https://x.com/raw_works/status/1999653061774721172)  2025-12-13T01:30Z [----] followers, [--] engagements


"this reminds me of the first time i found out about the mortal kombat cheat codes. .except now you can literally make anything you want. I like to make sure that I'm making some forward progress on every one of my active projects each day even when I'm too busy to spend real mental bandwidth on all of them every single day. So I've come up with a few prompts that I use a lot with the agents so they're always https://t.co/YUFEop90Do I like to make sure that I'm making some forward progress on every one of my active projects each day even when I'm too busy to spend real mental bandwidth on all"  
[X Link](https://x.com/raw_works/status/2000019742216650967)  2025-12-14T01:47Z [----] followers, [---] engagements


"@steipete i do everything with cli first before i dare to add human ui. despite all the various browser use tricks the feedback cycle grinds to a painfully slow pace as soon as there is a gui"  
[X Link](https://x.com/raw_works/status/2004643882055995602)  2025-12-26T20:02Z [----] followers, [--] engagements


"@DanielleFong have you tried gas town yet super fun"  
[X Link](https://x.com/raw_works/status/2008199163767570880)  2026-01-05T15:29Z [----] followers, [--] engagements


"@lateinteraction github gist or it didn't happen ;)"  
[X Link](https://x.com/raw_works/status/2008605241579893180)  2026-01-06T18:23Z [----] followers, [---] engagements


"@steipete i had to manually kill my clawdbot because of this bleeding heartbeat so sad"  
[X Link](https://x.com/raw_works/status/2009764069633343983)  2026-01-09T23:07Z [----] followers, [---] engagements


"@MaximeRivest different api keys vs oauth keys and very different pricing"  
[X Link](https://x.com/raw_works/status/2009765932961296725)  2026-01-09T23:15Z [----] followers, [--] engagements


"🤤📿 Introducing "Ralph Eats Beads" - an opinionated Ralph Wiggum implementation based on Claude Code beads and tmux"  
[X Link](https://x.com/raw_works/status/2012209308801364327)  2026-01-16T17:04Z [----] followers, [---] engagements


"How it works: Have your agent convert your plan into bd (beads) Ralph spawns parallel Claude agents in tmux Each agent implements ONE issue commits closes A watchdog kills stuck agents Loop until done 🔗 https://github.com/rawwerks/ralph-eats-beads https://github.com/rawwerks/ralph-eats-beads"  
[X Link](https://x.com/raw_works/status/2012209311879987583)  2026-01-16T17:04Z [----] followers, [---] engagements


"Install in Claude Code: /plugin marketplace add rawwerks/ralph-eats-beads /plugin install ralph-eats-beads@ralph-eats-beads"  
[X Link](https://x.com/raw_works/status/2012209314333614460)  2026-01-16T17:04Z [----] followers, [---] engagements


"This is "Gas Town Lite": more powerful than vanilla Ralph but faster to get started than Gas Town"  
[X Link](https://x.com/raw_works/status/2012209316250480895)  2026-01-16T17:04Z [----] followers, [--] engagements


"Built on @steveyegge's bd (beads) and inspired by @ghuntley's Ralph Wiggum loop concept"  
[X Link](https://x.com/raw_works/status/2012209318557331571)  2026-01-16T17:04Z [----] followers, [---] engagements


"@doodlestein i bought them about a year ago but never figured out the crappy software. with latest opus/codex - that shouldn't be a problem any more"  
[X Link](https://x.com/raw_works/status/2012641638766596181)  2026-01-17T21:42Z [----] followers, [---] engagements


"@doodlestein please do i bought a stream deck at your recommendation. i was able to get claude code to see the app and make a markdown table of the layout.but got distracted"  
[X Link](https://x.com/raw_works/status/2013069715199955319)  2026-01-19T02:03Z [----] followers, [--] engagements


"@dhh i made a fizzy cli for fun: err i guess i should say that claude code made it while i had fun. http://github.com/rawwerks/fizzy-cli http://github.com/rawwerks/fizzy-cli"  
[X Link](https://x.com/raw_works/status/2013084817944895496)  2026-01-19T03:03Z [----] followers, [--] engagements


"the man is simply asking for an easier way to pay anthropic for [--] max subscriptions so he can give away all of this claude code output away for free as open source.seems pretty reasonable to me. he should be the ICP for the claude code team - he literally cannot get enough @bcherny I really hope you will read this with an open mind and take it in the positive spirit it was intended; I say this as one of your biggest fans and supporters. I'm deeply grateful for the product you created which has changed my life profoundly. This one change to Claude @bcherny I really hope you will read this"  
[X Link](https://x.com/raw_works/status/2013404559670931781)  2026-01-20T00:13Z [----] followers, [---] engagements


"@lateinteraction it's funny the first thing i did today after seeing rlm now in dspy was to figure out how to spin up my own "semi-private" 8xH100 running glm [---] on modal - because i essentially wanted a version of @polymorfaisal 's use case but without the big labs logging my_entire_life"  
[X Link](https://x.com/raw_works/status/2013805185932370093)  2026-01-21T02:45Z [----] followers, [----] engagements


"too tired to make a big flashy post but for fun today my agents and i made a cli for rlm cc: @a1zhang @lateinteraction for example: rlm ask -q "Summarize this page" --json https://www.anthropic.com/constitution https://github.com/rawwerks/rlm-cli https://www.anthropic.com/constitution https://github.com/rawwerks/rlm-cli"  
[X Link](https://x.com/raw_works/status/2014160751867428904)  2026-01-22T02:18Z [----] followers, [----] engagements


"@ChiragCX @gooby_esq there are a few different ones i've found.but trying to make mine "the best". is mcp but i think clis are better than mcps appears to be a claude code implementation of rlm. https://github.com/brainqub3/claude_code_RLM https://github.com/Hmbown/aleph https://github.com/brainqub3/claude_code_RLM https://github.com/Hmbown/aleph"  
[X Link](https://x.com/raw_works/status/2014357754274619527)  2026-01-22T15:21Z [----] followers, [--] engagements


"@zeeg i can think of at least 3: @GeoffreyHuntley @Steve_Yegge @doodlestein"  
[X Link](https://x.com/raw_works/status/2014377994312167531)  2026-01-22T16:41Z [----] followers, [---] engagements


"@doodlestein @zeeg @GeoffreyHuntley @Steve_Yegge beads has tens of thousands of users take a look at jeff's stuff - it is quite impressive"  
[X Link](https://x.com/raw_works/status/2014379547781632485)  2026-01-22T16:48Z [----] followers, [--] engagements


"@tiagoefreitas @a1zhang no what i did is make a cli wrapper around rlm. but there is someone who attempted to implement rlm directly in claude code: https://github.com/brainqub3/claude_code_RLM https://github.com/brainqub3/claude_code_RLM"  
[X Link](https://x.com/raw_works/status/2014514100470939878)  2026-01-23T01:42Z [----] followers, [--] engagements


"here's what i was able to get cc to reveal: # Claude Code Tasks and Swarms **Status**: Undocumented features as of January [----]. This document is based on system prompt analysis not official docs. ## Overview Tasks are a new primitive in Claude Code that replace/evolve the old TodoWrite tool. They're designed for: - Tracking complex multi-step work - Cross-session collaboration - Multi-agent (swarm) coordination ## Task Tools Tool Purpose --------------- TaskCreate Create a task with subject description activeForm  TaskGet Retrieve full task details by ID  TaskUpdate Update status"  
[X Link](https://x.com/raw_works/status/2015084794229538979)  2026-01-24T15:30Z [----] followers, [---] engagements


"@mikedemarais great question (but wrong platform for it.) [--]. ask cc to install clawdbot for you. [--]. ask clawdbot this exact question. [--]. margaritas"  
[X Link](https://x.com/raw_works/status/2015086666805891438)  2026-01-24T15:37Z [----] followers, [---] engagements


"@badlogicgames @moltbot plz bring back geocities @moltbot"  
[X Link](https://x.com/raw_works/status/2015471938894770358)  2026-01-25T17:08Z [----] followers, [---] engagements


"working on adding explicit "user approval" to open prose mostly so i can stop cursing at @moltbot curious for any feedback on this approach: btw - awesome work @irl_danB open prose is super powerful https://github.com/rawwerks/prose https://github.com/rawwerks/prose"  
[X Link](https://x.com/raw_works/status/2015484225277854016)  2026-01-25T17:57Z [----] followers, [---] engagements


"@kristianfreeman @moltbot but does it produce your techno (please add techno skill to clawdhub)"  
[X Link](https://x.com/raw_works/status/2015488588436095128)  2026-01-25T18:15Z [----] followers, [---] engagements


"thanks i didn't see inputs before i started this and i don't think that what it did is entirely redundant. the thought process was that there should be an enforced trust boundary that the ai running the program cannot get past on its own. this doesn't actually have to be "human" in the loop could be another agent that is overseeing prose.but either way i tried to add some "extra determinism" via gates. what i've done so far (still actively working now): approve gates - checkpoints that the executing agent cannot self-resolve approve deploy_gate: prompt: "Deploy these changes" allow: "user""  
[X Link](https://x.com/raw_works/status/2015495544408592499)  2026-01-25T18:42Z [----] followers, [--] engagements


"@thekitze clawdbot is really good at obsidian if you give it qmd from @tobi = https://github.com/tobi/qmd https://github.com/tobi/qmd"  
[X Link](https://x.com/raw_works/status/2015499511334920587)  2026-01-25T18:58Z [----] followers, [---] engagements


"@0xMerth @moltbot @doodlestein whoops i should have kept scrolling. https://x.com/doodlestein/status/2015813204056674604s=20 @DanielMiessler Use ACIP: https://t.co/RfsIkhJJog https://x.com/doodlestein/status/2015813204056674604s=20 @DanielMiessler Use ACIP: https://t.co/RfsIkhJJog"  
[X Link](https://x.com/raw_works/status/2015829719275466917)  2026-01-26T16:50Z [----] followers, [---] engagements


"snow day trying to force myself to use openprose for everything. amazing tag-team job from codex cli and claude code on implementing this very helpful new primitive. exec a new OpenProse primitive that runs shell commands directly through the VM without spawning a subagent. # Before: spawns an Opus subagent (15s 5k tokens) just to type a command let status = session "Run git status" # After: VM runs it directly (1s [--] tokens) let status = exec "git status" Why it matters Every computation in OpenProse previously required a session which means spawning a full LLM subagent via the Task tool. For"  
[X Link](https://x.com/raw_works/status/2015873040786108805)  2026-01-26T19:42Z [----] followers, [---] engagements


"forgot to cc @moltbot and @irl_danB"  
[X Link](https://x.com/raw_works/status/2015873581402460634)  2026-01-26T19:44Z [----] followers, [--] engagements


"out of curiosity after reading this i started benchmarking rlm and dspy.rlm on longmemeval tl;dr - i think i might have a new "sota memory system" by the end of the day. cc @DSPyOSS @a1zhang @lateinteraction 🚨 Announcing a new SOTA memory system Observational Memory (OM) available in latest @mastra version now. It achieves the highest scores ever recorded on LongMemEval (gpt-4o 84.2% gpt-5-mini 94.9%) No RAG no graphs no input based retrieval just a simple constantly evolving https://t.co/Xk9Y6JtR63 🚨 Announcing a new SOTA memory system Observational Memory (OM) available in latest @mastra"  
[X Link](https://x.com/raw_works/status/2021303970929479795)  2026-02-10T19:23Z [----] followers, 45.9K engagements


"today i'm releasing "ypi" a fully recursive coding agent. imagine rlms (@a1zhang & @lateinteraction) and pi (@badlogicgames) had a baby. bunx ypi "Prove to me you are truly recursive." (links below) hat tip to @GeoffreyHuntley for turning me onto jj"  
[X Link](https://x.com/raw_works/status/2022010439446802814)  2026-02-12T18:10Z [----] followers, 20.7K engagements


"@ovsky4 i wasn't able to get the "true essence of RLM" working as a pure extension but i bet it is possible"  
[X Link](https://x.com/raw_works/status/2022353618557632653)  2026-02-13T16:54Z [----] followers, [--] engagements


"@DanielleFong cass by @doodlestein can do something similar. index and semantically query all coding agent convos - then you can resume the id or just cherry pick the stuff you need into the current agent = https://github.com/Dicklesworthstone/coding_agent_session_search https://github.com/Dicklesworthstone/coding_agent_session_search"  
[X Link](https://x.com/raw_works/status/2013088587957375258)  2026-01-19T03:18Z [----] followers, [--] engagements


"Study hard what interests you the most in the most undisciplined irreverent and original manner possible. Richard Feynmann I am currently studying: ralph wiggum rlms dspy gastown & related agent orchestrators agent-native apps and orgs How about you"  
[X Link](https://x.com/raw_works/status/2014360452784869611)  2026-01-22T15:32Z [----] followers, [---] engagements


"@a1zhang thank you for clarifying this i had a similar debate with gpt [---] pro about "ralph wiggum vs. rlm" (fwiw - cc pretty intuitively understands how to use rlm when it is packaged up nicely: ) https://github.com/rawwerks/rlm-cli https://github.com/rawwerks/rlm-cli"  
[X Link](https://x.com/raw_works/status/2014375592150327571)  2026-01-22T16:32Z [----] followers, [---] engagements


"@MiniMax_AI @steipete wow clawbot is crossing the chasm"  
[X Link](https://x.com/raw_works/status/2014445832649441422)  2026-01-22T21:11Z [----] followers, [----] engagements


"i had to get clawdbot to give me a full analysis of how its message queue gets actually turned into the message array for the context window. two helpful things: - tell it to write things to files - install "cass" for fast searching of all prior coding agent convos: https://github.com/Dicklesworthstone/coding_agent_session_search https://github.com/Dicklesworthstone/coding_agent_session_search"  
[X Link](https://x.com/raw_works/status/2014726342806225148)  2026-01-23T15:46Z [----] followers, [---] engagements


"@irl_danB @moltbot nice yeah it's funny to see everyone this morning so excited to get rid of their kids ;)"  
[X Link](https://x.com/raw_works/status/2016183288872751331)  2026-01-27T16:15Z [----] followers, [--] engagements


"wow claude code got a bit overconfident while building a new coinbase api skill and literally placed a limit sell order for all of my btc (and that was without any prompt injection) i guess it's called yolo mode for a reason"  
[X Link](https://x.com/raw_works/status/2016199603696128006)  2026-01-27T17:20Z [----] followers, [--] engagements


"@a1zhang i will if you want didn't want to spam with unsolicited stuff.been making a lot of changes to rlm. i also implemented depth1 with a bunch of helpful cost/time/repl budgeting stuff - do you want that too"  
[X Link](https://x.com/raw_works/status/2016583272000544774)  2026-01-28T18:44Z [----] followers, [---] engagements


"@kloss_xyz @karpathy @steipete @gregisenberg @rileybrown @corbin_braun @jackfriks @EXM7777 @eptwts you forgot @doodlestein @GeoffreyHuntley @Steve_Yegge @badlogicgames @mitsuhiko @lateinteraction @a1zhang @swyx and so many more"  
[X Link](https://x.com/raw_works/status/2016620945478181217)  2026-01-28T21:14Z [----] followers, [---] engagements


"inspired by last week's research report from @gao_jude at @vercel - "AGENTS.md outperforms skills in our agent evals" this weekend my agents worked hard to make dirpack. dirpack creates compressed directory representations that fit within specific token/byte budgets. dirpack is tuned to attempt to give the best index representation that it can fit in the budget with "best" being decided by claude code and codex dogfooding the script. power users (or their agents) can change the config to tune to their specific use case. initial results suggest that dirpack offers a very fast way to orient"  
[X Link](https://x.com/raw_works/status/2018353444042543537)  2026-02-02T15:58Z [----] followers, [--] engagements


"as much as i love @openclaw and also running pi on termux there is something very special about a simple web ui to my local machine's claude code that i can then use from anywhere and any device via @Tailscale (frankly this is what claude code mobile/web should just be) I reverse-engineered Claude Code's binary. Found a flag they hid from --help. --sdk-url Enable it and the terminal disappears. The CLI becomes a WebSocket client. We built a server to catch the connection. Added a React UI on top. Now I run Claude Code from my browser. From https://t.co/GHCLStJ4FD I reverse-engineered Claude"  
[X Link](https://x.com/raw_works/status/2021051381076066447)  2026-02-10T02:39Z [----] followers, [----] engagements


"i had a chance to ask @a1zhang about coding agents x rlms on @askalphaxiv - see his response here: in case folks are curious - i've been trying "both approaches" 1) rlm as a cli tool here: and 2) tonight i'm working on an extension to @badlogicgames 's pi that calls rlm under the hood instead of llm. http://github.com/rawwerks/rlm-cli https://youtu.be/6Dr3SUmHFcosi=YNs4dTvAvJE53nYZ&t=939 http://github.com/rawwerks/rlm-cli https://youtu.be/6Dr3SUmHFcosi=YNs4dTvAvJE53nYZ&t=939"  
[X Link](https://x.com/raw_works/status/2021055582505963559)  2026-02-10T02:56Z [----] followers, [----] engagements


"@odd_joel @openclaw @Tailscale nice i'll have to check that out. was trying et via termius for a bit"  
[X Link](https://x.com/raw_works/status/2021252388170957192)  2026-02-10T15:58Z [----] followers, [--] engagements


"@DanielleFong what does the lightsaber agent do"  
[X Link](https://x.com/raw_works/status/2021281455809130657)  2026-02-10T17:53Z [----] followers, [--] engagements


"preliminary results: dspy.RLM + Gemini [--] Flash: 87.2% on LongMemEval_S Beats Mastra Observational Memory + GPT-4o (84.23%) at 1/5th the per-token cost"  
[X Link](https://x.com/raw_works/status/2021303973471268923)  2026-02-10T19:23Z [----] followers, [----] engagements


"The same Gemini [--] Flash scored 57.9% with the standalone rlm package. Switching to dspy.RLM structured inputs iterative REPL SUBMIT() added +29 points. The scaffolding was everything"  
[X Link](https://x.com/raw_works/status/2021304330003824997)  2026-02-10T19:24Z [----] followers, [----] engagements


"- standalone rlm + Gemini [--] Flash: 58.0% - standalone rlm + Gemini [--] Pro: 87.0% - dspy.RLM + Flash: 87.2% - Mastra OM + GPT-4o: 84.2%"  
[X Link](https://x.com/raw_works/status/2021307467494289911)  2026-02-10T19:37Z [----] followers, [----] engagements


"@huntermbown i'm convinced that having $100k in vertex credits and a tier [--] rate limit makes gemini waaaay more rlm-friendly (for me)"  
[X Link](https://x.com/raw_works/status/2021307947406504030)  2026-02-10T19:39Z [----] followers, [--] engagements


"and you will probably want my time and cost and GEPA helpers around the DSPy.rlm = https://github.com/rawwerks/dspy/tree/feat/rlm-multimodal-media-support https://github.com/rawwerks/dspy/tree/feat/rlm-multimodal-media-support"  
[X Link](https://x.com/raw_works/status/2021315145994056116)  2026-02-10T20:07Z [----] followers, [----] engagements


"cc @DSPyOSS https://github.com/stanfordnlp/dspy/issues/9289 https://github.com/stanfordnlp/dspy/issues/9289"  
[X Link](https://x.com/raw_works/status/2021316718782517369)  2026-02-10T20:13Z [----] followers, [----] engagements


"dspy.RLM + Pro final: 89.4% on 500/500 Mastra OM + GPT-4o: 84.2% (reported) standalone rlms + Flash: 58.0% (500/500) standalone rlms + Pro: 87.0% (500/500) dspy.RLM + Flash: 87.2% (500/500) dspy.RLM + Pro: 89.4% (500/500) new best Breakdown by category: - single-session-assistant: 98.2% - single-session-user: 95.7% - temporal-reasoning: 91.0% - knowledge-update: 89.7% - single-session-preference: 90.0% - multi-session: 80.5% still the hardest https://twitter.com/i/web/status/2021323835413348845 https://twitter.com/i/web/status/2021323835413348845"  
[X Link](https://x.com/raw_works/status/2021323835413348845)  2026-02-10T20:42Z [----] followers, [----] engagements


"@kmad @DSPyOSS @isaacbmiller1 i'm just trying to follow the contribution guidelines by making an issue first.but honestly with the speed and power of claude code at this point i'm fully comfortable maintaining my own fork if @DSPyOSS maintainers don't like these improvements"  
[X Link](https://x.com/raw_works/status/2021398434620842121)  2026-02-11T01:38Z [----] followers, [--] engagements


"@kmad @DSPyOSS @isaacbmiller1 will have the agents take a look at your approach shortly"  
[X Link](https://x.com/raw_works/status/2021398675969409320)  2026-02-11T01:39Z [----] followers, [--] engagements


"here's what cc says (below) q: @isaacbmiller1 is dspy.Type here to stay Good call. Here's what I take away from PR #9283: Key Lessons from the DataFrame PR [--]. They're building a dspy.Type protocol for sandbox injection and it's generic. The PR introduces three methods on dspy.Type subclasses that any custom type can implement: - sandbox_setup() import statements (e.g. import pandas as pd) - to_sandbox() serialize to bytes for file-based injection - sandbox_assignment(var_name data_expr) code string to reconstruct the object in the sandbox This is a much cleaner pattern than what our branch"  
[X Link](https://x.com/raw_works/status/2021399974592881135)  2026-02-11T01:44Z [----] followers, [--] engagements


"@kmad @DSPyOSS @isaacbmiller1 let's have both of our agents call @lateinteraction and get all of this worked out"  
[X Link](https://x.com/raw_works/status/2021401812457394429)  2026-02-11T01:52Z [----] followers, [--] engagements


"unfortunately i could not get the @DSPyOSS team to care about "dspy for cli" in november. fortunately everything needed for this idea is now implemented for DSPy.rlm because rlm is basically a coding agent. and so now "dspy for cli" is refactored to follow the rlm approach: try running dspy on claude code codex pi etc https://github.com/stanfordnlp/dspy/issues/9034 here's my pitch: CLIs for DSPy use and optimize any coding agent with the power of DSPy. would appreciate any feedback on this idea from the @DSPyOSS peeps and especially @lateinteraction @MaximeRivest @a1zhang @LakshyAAAgrawal"  
[X Link](https://x.com/raw_works/status/2021403449846288631)  2026-02-11T01:58Z [----] followers, [----] engagements


"@mikehostetler @DSPyOSS sure which part though i don't know anything about elixir but i'm not sure the pieces i added to dspy.rlm would really help in a completely different language"  
[X Link](https://x.com/raw_works/status/2021406994750652537)  2026-02-11T02:12Z [----] followers, [---] engagements


"@DSPyOSS 91.2% at 250/500. Holding steady above 91%. Predictions are at 216+ and judge is catching up"  
[X Link](https://x.com/raw_works/status/2021407670197252300)  2026-02-11T02:15Z [----] followers, [---] engagements


"@kmad @DSPyOSS @isaacbmiller1 @lateinteraction hey @kmad i just went ahead and had claude code do it: please just yell at my agent if necessary. https://github.com/rawwerks/dspy/tree/feat/rlm-media-types-protocol https://github.com/stanfordnlp/dspy/pull/9283#issuecomment-3881740291 https://github.com/rawwerks/dspy/tree/feat/rlm-media-types-protocol https://github.com/stanfordnlp/dspy/pull/9283#issuecomment-3881740291"  
[X Link](https://x.com/raw_works/status/2021408951536058850)  2026-02-11T02:20Z [----] followers, [--] engagements


"ok well i don't think i can say fully SOTA before bed but not bad for one day of work on a benchmark i just learned about this morning. tl;dr - with DSPy.rlm you can get Gemini [--] Flash Preview to 89.8% on LongMemEval without any special preprocessing "memory system". this score is in the top [--] of the leaderboard and with Flash it is very inexpensive. i'll try to write something proper up tomorrow in the meantime some notes from my agent about what worked and what didn't. out of curiosity after reading this i started benchmarking rlm and dspy.rlm on longmemeval tl;dr - i think i might have a"  
[X Link](https://x.com/raw_works/status/2021412999932612648)  2026-02-11T02:36Z [----] followers, [----] engagements


"From the leaderboard: - dspy.RLM Flash baseline (no tools): 87.2% - dspy.RLM Flash + tools + prompt (exp 004): 89.2% - dspy.RLM Flash + obs memory + tools (exp 006): 89.8% The tools (search_sessions grep_sessions get_session) + structured prompt added +2.6pp over plain dspy.RLM. That's [--] more questions correct out of [---] going from [---] to [---]. So the full picture: - Plain dspy.RLM (Flash no tools) 87.2% (baseline) - + tools + prompt engineering 89.2% (+2.0pp) - + typed Observation output + temporal grounding 89.8% (+0.6pp more) - + GEPA prompt optimization 87.8% (-2.0pp hurt) - + swap to Pro"  
[X Link](https://x.com/raw_works/status/2021413002939871552)  2026-02-11T02:36Z [----] followers, [----] engagements


"@gooby_esq @OpenAIDevs did you do as a tool or with rlm instead of llm"  
[X Link](https://x.com/raw_works/status/2021413513101762573)  2026-02-11T02:38Z [----] followers, [---] engagements


"@tiagoefreitas for my rlm-cli i made a claude plugin/skill: https://github.com/rawwerks/rlm-cli/tree/master/.claude-plugin https://github.com/rawwerks/rlm-cli/tree/master/.claude-plugin"  
[X Link](https://x.com/raw_works/status/2021612109356806217)  2026-02-11T15:47Z [----] followers, [--] engagements

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