Dark | Light
# ![@svpino Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::18737039.png) @svpino Santiago

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

### Engagements: [-------] [#](/creator/twitter::18737039/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::18737039/c:line/m:interactions.svg)

- [--] Week [---------] -9.10%
- [--] Month [---------] +8.90%
- [--] Months [----------] -40%
- [--] Year [----------] -52%

### Mentions: [--] [#](/creator/twitter::18737039/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::18737039/c:line/m:posts_active.svg)

- [--] Week [---] +27%
- [--] Month [---] +55%
- [--] Months [---] +41%
- [--] Year [-----] -8.60%

### Followers: [-------] [#](/creator/twitter::18737039/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::18737039/c:line/m:followers.svg)

- [--] Week [-------] +0.21%
- [--] Month [-------] +0.78%
- [--] Months [-------] +3.70%
- [--] Year [-------] +11%

### CreatorRank: [------] [#](/creator/twitter::18737039/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::18737039/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[travel destinations](/list/travel-destinations)  #1909 [technology brands](/list/technology-brands)  30% [social networks](/list/social-networks)  6% [finance](/list/finance)  6% [stocks](/list/stocks)  3% [products](/list/products)  1% [countries](/list/countries)  1%

**Social topic influence**
[ai](/topic/ai) #1786, [code](/topic/code) #419, [claude code](/topic/claude-code) #48, [if you](/topic/if-you) 10%, [build](/topic/build) #1015, [this is](/topic/this-is) 8%, [model](/topic/model) #1840, [how to](/topic/how-to) 7%, [agentic](/topic/agentic) #97, [engineering](/topic/engineering) #106

**Top accounts mentioned or mentioned by**
[@lfuckingg](/creator/undefined) [@grok](/creator/undefined) [@mingtakaivo](/creator/undefined) [@gaekingai](/creator/undefined) [@unclejai](/creator/undefined) [@ghidorah_x](/creator/undefined) [@amanrawatamg](/creator/undefined) [@tacodevs](/creator/undefined) [@nabbilkhan](/creator/undefined) [@wildpinesai](/creator/undefined) [@pamelafox](/creator/undefined) [@ejae_dev](/creator/undefined) [@degenduke0](/creator/undefined) [@saen_dev](/creator/undefined) [@cliwant_patrick](/creator/undefined) [@vlelyavin](/creator/undefined) [@rahul18kaushik](/creator/undefined) [@8snotreal](/creator/undefined) [@yungenfant](/creator/undefined) [@joshua_m_storm](/creator/undefined)

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

"Most companies right now: - No automated tests - No code review process - No CI/CD pipelines - Poor secret management - No dataset versioning - Production workflows run from spreadsheets - No rollback plans - No integration tests These aren't just some weird companies. They're everywhere The market for people who can fix these fundamentals is massive. https://twitter.com/i/web/status/2017591212342128958 https://twitter.com/i/web/status/2017591212342128958"  
[X Link](https://x.com/svpino/status/2017591212342128958)  2026-01-31T13:30Z 437.6K followers, 273.2K engagements


"You can set up moltbot (now openclaw) on Emergent in under [--] minutes. You don't need to buy a Mac mini use a terminal or worry about configuration files. Here's what you can do: [--]. Go to Emergent's homepage [--]. Click on the MoltBot option [--]. A prompt will automatically show up [--]. Click next That's it. You can then connect your WhatsApp or Telegram account to start using the bot immediately. This turns [--] hours of installation and configuration into [--] minutes. But even more important: This will help you keep things secure and away from your computer. MoltBot is fun and powerful but I do not"  
[X Link](https://x.com/svpino/status/2017695040529961145)  2026-01-31T20:22Z 437.6K followers, 83.1K engagements


"This is literally the fastest way to install OpenClaw (MoltBot). This video will show you how to do it step by step. You don't need to buy a Mac Mini. We'll install it in @digitalocean. I've installed this [--] different times already and this is the fastest way I've found"  
[X Link](https://x.com/svpino/status/2019048736203456710)  2026-02-04T14:01Z 437.6K followers, 239K engagements


"Claude Code /insights command is insanely useful Do this: [--]. Run the /insights command [--]. Find the generated report file [--]. Attach it to a Claude session Prompt: Read the attached report and create a bullet list with [--] lessons to improve my use of Claude Code. Mind blown"  
[X Link](https://x.com/svpino/status/2021578475166146615)  2026-02-11T13:34Z 437.6K followers, 34.8K engagements


"MiniMax M2.1 is really good for writing code. It's open source and you can connect Visual Studio Code to it and start coding right away. The model is especially good for those who don't use Python as a primary language. It's currently the best multilingual model out there. The model hits 72.5% on SWE-Multilingual and 88.6% on VIBE-Bench (best) This means the model is also really good at performing real engineering work and does very well at agentic tasks: Long tool chains Structured agent scaffolding Following repo-level rules and slash commands Working inside existing agent frameworks"  
[X Link](https://x.com/svpino/status/2015877777732600082)  2026-01-26T20:01Z 437.6K followers, 48.3K engagements


"I'm giving a free talk at Sonar Summit in March. I'll focus on Claude Code and share dozens of tips for using it more effectively. My talk will be FREE and ONLINE. Everyone can attend. RSVP using this link: https://fandf.co/406EbDf https://fandf.co/406EbDf"  
[X Link](https://x.com/svpino/status/2021578487048568889)  2026-02-11T13:34Z 437.6K followers, [----] engagements


"By the way the insights report that Claude Code generates is packed with useful information. You can simply read the report yourself and it will give you a bunch of ideas that you can implement to improve it. There's a lot of information so I like to use Claude to distill what's important for me. Some examples I got from my report (these might not make sense for everyone because they are specific to my use cases): [--]. The report recommended including my file structure in CLAUDE .md [--]. Name the tool in my prompts when finding business information [--]. Add a PreToolUse hook to run a function 4."  
[X Link](https://x.com/svpino/status/2021578499245641778)  2026-02-11T13:34Z 437.6K followers, [----] engagements


"sOfTwArE eNgInEeRiNg iS dEaD you have to be mentally challenged to think your mom will use ai to build a grocery app whenever she needs one"  
[X Link](https://x.com/svpino/status/2021996960845377826)  2026-02-12T17:16Z 437.6K followers, 304.4K engagements


"I've been running 4-6 coding agents every single day for months now: I have [--] - [--] instances of Claude Code. I have Copilot. I have Warp. I have Jules. I'm figuring this out as I go but it's getting really hard to keep track of everything that's happening. 90% of the time I can't remember which agent is working on what. I can't keep context in my head and I forget to check the agent that finished [--] minutes ago. Models aren't the bottleneck anymore. I am. I think there's one solution for this: https://twitter.com/i/web/status/2022043267945443337"  
[X Link](https://x.com/svpino/status/2022043267945443337)  2026-02-12T20:20Z 437.6K followers, 20.4K engagements


"engineering team shipping dark mode during a DDoS attack BREAKING: House Republicans announce investigation into Bad Bunnys Super Bowl performance alleging the Spanish lyrics were inappropriate for television. BREAKING: House Republicans announce investigation into Bad Bunnys Super Bowl performance alleging the Spanish lyrics were inappropriate for television"  
[X Link](https://x.com/svpino/status/2022332413159125379)  2026-02-13T15:29Z 437.6K followers, 11K engagements


"Honestly only nerds care about coding agents. What your mom friends and neighbors want is *something* that does real things for them. Lindy is a pretty dope assistant. Nothing to install and nothing to configure. You don't need a technical cousin to run this for you. Here is a quick list of what it does: It checks your email It can reply back using your voice It schedules meetings It pulls context from past conversations before meetings It takes notes during meetings It tells you what's important It can draft and send follow-ups It can text you on iMessage when it needs you. This is the type"  
[X Link](https://x.com/svpino/status/2022414300401533348)  2026-02-13T20:55Z 437.6K followers, 15.1K engagements


"@BelieveOnJesus The fact that AI already replaced 100% of your coding tasks says a lot about the complexity of the work you were doing before"  
[X Link](https://x.com/svpino/status/2022684515269820621)  2026-02-14T14:49Z 437.6K followers, [----] engagements


"@_flashman_harry Sure thing. We are also colonizing Mars. sorry I meant the Moon and we will have 100% self-driving cars by the end of 2020"  
[X Link](https://x.com/svpino/status/2022690675654054172)  2026-02-14T15:13Z 437.6K followers, [----] engagements


"You are so close of getting it Its so hard to contrast Boris take with Darios claim that: software engineering is fully automated by [----]. You might have the 0.01% engineer left over but that's it the job is finished. Why does Dario keep making such insane claims Its so hard to contrast Boris take with Darios claim that: software engineering is fully automated by [----]. You might have the 0.01% engineer left over but that's it the job is finished. Why does Dario keep making such insane claims"  
[X Link](https://x.com/svpino/status/2022875414797136255)  2026-02-15T03:27Z 437.6K followers, 27.7K engagements


"11 ways ChatGPT saves me hours of work every day and why you'll never outcompete those who use AI effectively. A list for those who write code: [--] of 16"  
[X Link](https://x.com/anyuser/status/1624392828154712064)  2023-02-11T13:00Z 437.7K followers, 2.1M engagements


"Another deep learning breakthrough: Deep TDA a new algorithm using self-supervised learning overcomes the limitations of traditional dimensionality reduction algorithms. t-SNE and UMAP have long been the favorites. Deep TDA might change that forever. Here are the details:"  
[X Link](https://x.com/svpino/status/1676561661891854338)  2023-07-05T12:00Z 436.9K followers, 1.1M engagements


"Best GPU prices you'll find out there: H100 GPU for $2.00/hr H200 GPU for $2.30/hr Even better: You can literally move all of your data to Nebius *for free* including a free PoC with support and get up to [--] months for free Nebius is collaborating with me on this post and their deal is just amazing This is insanely cheap and you should check it out if you are looking to fine-tune models run experiments or do some production testing. Check this out: https://bit.ly/46rANG7 https://bit.ly/46rANG7"  
[X Link](https://x.com/svpino/status/1970143734408101970)  2025-09-22T15:10Z 436.9K followers, 59.3K engagements


"I still think that prompt engineering is not a career. I still think companies will not start hiring prompt engineers anytime soon. I still think about prompt engineering as a skill not a role. Here is a counterintuitive lesson from building agents: The prompt is often more important than the system's architecture. I've always been obsessed with breaking down my system into the proper number of agents and assigning responsibilities correctly but the biggest gains in Here is a counterintuitive lesson from building agents: The prompt is often more important than the system's architecture. I've"  
[X Link](https://x.com/svpino/status/1971559145431257323)  2025-09-26T12:55Z 437K followers, 15.8K engagements


"Building software is only 10% about writing code and 90% about thinking what code to write why you should write it and how to do it well. You can delegate the code-writing part to an LLM but so far they have shown they cant handle the thinking part. Its called vibe-coding and not vibe-thinking for a reason. Vibe-coders cant build the type of software $400000/year salaries pay today. If you believe companies pay that money to build vibe-code-able apps you should crack a book and get off Twitter. Vibe-coders cant build the type of software $400000/year salaries pay today. If you believe"  
[X Link](https://x.com/svpino/status/1977396812999688371)  2025-10-12T15:31Z 437.6K followers, 168.1K engagements


"Ive been building software for [--] years more than [--] professionally. I understand architecture principles and best practices. No you cant vibe-code Salesforce SAP or any other complex enterprise app. @svpino @Rifadm816 If you understand architecture principles and best practices then yes absolutely. @svpino @Rifadm816 If you understand architecture principles and best practices then yes absolutely"  
[X Link](https://x.com/svpino/status/1993762257080033721)  2025-11-26T19:22Z 437.6K followers, 374.2K engagements


"I don't believe you can vibe-code complex applications. Put another way your ability to vibe-code software is inversely proportional to its complexity. Vibe-coding is like going into the "Minotaur Labyrinth" without Theseus' thread: you'll invariably get lost. More accurately you'll be stuck in an endless loop going in circles with the model: Things that used to work stop working Your code starts bloating Solutions become over-engineered And all of this will happen while working with popular tech where models have plenty of training. Try to step outside the beaten path (use anything a bit"  
[X Link](https://x.com/svpino/status/1994031036720926881)  2025-11-27T13:10Z 437.7K followers, 78.5K engagements


"There is a massive opportunity right now for anyone who can walk into a company and automate the grunt work they do every day. It doesn't matter what programming language you use or libraries you like or whether you know what micro-services do or if you like Claude Code or Cursor. No one cares about any of that. If you can identify a problem and convince them you can solve it you'll make a killing"  
[X Link](https://x.com/svpino/status/1994402155906859223)  2025-11-28T13:45Z 436.9K followers, 50.2K engagements


"68% of developers say they spend more time debugging AI code than writing new code. (According to a survey of [---] software engineering leaders and practitioners.) AI coding assistants help us write code faster but they increase the amount of insecure and broken code we ship"  
[X Link](https://x.com/svpino/status/2013265452294893660)  2026-01-19T15:01Z 437.6K followers, 36.9K engagements


"All these [--] models are now available: [--]. Qwen3-Next-80B-A3B-Thinking [--]. DeepSeek-V3.2 [--]. MiniMax M2.1 [--]. Nemotron-3-Nano-30B-A3B Open source AI is unstoppable. Excited to announce two new models added to Token Factory First: @nvidia Nemotron-3-Nano-30B-A3B a compact MoE (30B params 3.6B active) for coding reasoning instruction-following tool-calling. Second: @Alibaba_Qwen Qwen3-Next-80B-A3B-Thinking a powerful MoE (80B params https://t.co/EIlt53FhHK Excited to announce two new models added to Token Factory First: @nvidia Nemotron-3-Nano-30B-A3B a compact MoE (30B params 3.6B active) for"  
[X Link](https://x.com/svpino/status/2014805847243948531)  2026-01-23T21:02Z 437.3K followers, 31.1K engagements


"Voice agents are awkward and everyone notices: You ask a question. The agent thinks. You wait. And wait. Nobody wants this. I'd rather talk to a person. If your model's response time is over 300ms you won't make it. Unfortunately most text-to-speech models can't get anywhere close to that. I want you to take a look at the latest model released by @inworld_ai: TTS-1.5. I built a simple voice agent using the model so you can see it in action and test it on your computer. You'll find the repository link below. The latency numbers of this model are wild: Max model under 250ms Mini model under"  
[X Link](https://x.com/svpino/status/2016182120994243032)  2026-01-27T16:10Z 437.7K followers, 69.2K engagements


"GitHub just released its Copilot SDK. You can now embed Copilot into any of your applications and make it much more powerful with just a few lines of code. I built a quick example here"  
[X Link](https://x.com/svpino/status/2016542131452883057)  2026-01-28T16:01Z 437.5K followers, 60.7K engagements


"I don't remember the last time I was so excited about something as I am right now about Claude Code"  
[X Link](https://x.com/svpino/status/2018316005513322624)  2026-02-02T13:30Z 437.3K followers, 51.1K engagements


"I've been trying to move from imperative prompting to declarative prompting. Imperative prompting: I tell the agent how to do something. Declarative prompting: I describe my goal to the agent. This was impossible before but since Opus [---] I'm letting models fill in the blanks more and more. I recorded a video using Quest + Opus [---]. This is a coding agent designed specifically for this purpose"  
[X Link](https://x.com/svpino/status/2018356346366919035)  2026-02-02T16:10Z 436.9K followers, 96.6K engagements


"So many people are shipping AI-generated code without even looking at it. It's terrifying AI models are improving but they still make a ton of mistakes. Now more than ever you get code that looks fine and passes a quick glance but underneath it doesn't work: race conditions broken contracts duplicated logic and any number of subtle issues that aren't obvious. This will come back to haunt us forever. The team behind Qodo just launched their [---] version to address this. https://twitter.com/i/web/status/2019140163499352527 https://twitter.com/i/web/status/2019140163499352527"  
[X Link](https://x.com/svpino/status/2019140163499352527)  2026-02-04T20:05Z 437K followers, 19.1K engagements


"Sam meet Sam. First the good part of the Anthropic ads: they are funny and I laughed. But I wonder why Anthropic would go for something so clearly dishonest. Our most important principle for ads says that we wont do exactly this; we would obviously never run ads in the way Anthropic First the good part of the Anthropic ads: they are funny and I laughed. But I wonder why Anthropic would go for something so clearly dishonest. Our most important principle for ads says that we wont do exactly this; we would obviously never run ads in the way Anthropic"  
[X Link](https://x.com/svpino/status/2019151214345138389)  2026-02-04T20:48Z 437.5K followers, 96K engagements


"There are so many people telling these amazing stories of how they built these great applications overnight using AI and how they are making so much money because of it And I think its all bullshit. Ive been using these models every day since they were 10x more dumb they are today and theres absolutely no chance non-devs are one-shotting anything of value without significant manual work. My honest opinion: people who claim to have easily "one shotted" something with AI probably work on *very* conventional use cases. Or have ready made templates of steps that need to be accomplished. Every"  
[X Link](https://x.com/svpino/status/2019199832825688337)  2026-02-05T00:02Z 437.6K followers, 94.4K engagements


"Opus [---] is so good that declarative prompting is now becoming possible for the first time. Here is how I usually think about prompting the model: "Rewrite this function using a hash map instead of nested loops." That's "imperative prompting." With the latest iteration you can now try this instead: "This function is too slow. Make it faster without changing what it returns." That's declarative prompting. This doesn't always work (the agent sometimes needs more context and details to not screw things up) but my success rate is much higher than ever before."  
[X Link](https://x.com/svpino/status/2019410949749559621)  2026-02-05T14:01Z 437.6K followers, 71.9K engagements


"Testing sucks across most companies I work with. I see this over and over again: Most teams don't write automated tests. Some of them do some formal manual testing. Some do have automated tests but nobody trusts them. Writing automated tests is hard. Maintaining them is even harder. Every UI change every new feature every refactor breaks something and someone has to go fix the tests. AI can help here. Look at the attached video"  
[X Link](https://x.com/svpino/status/2019472955139584447)  2026-02-05T18:07Z 437.3K followers, 13.4K engagements


"Virtual assistants should be scared @genspark_ai just hit $155M in ARR in only [--] months. This is an all-in-one AI workspace that replaces multiple tools: Slides Design Images Data Research And it's all inside one interface. Here's the crazy part: It's $19.99/month and you get access to the best AI models in the world + a specialized agents that do the work for you"  
[X Link](https://x.com/svpino/status/2019520168364503411)  2026-02-05T21:15Z 437.3K followers, [----] engagements


"You can also ask Genspark to build a website similar to another. Same style. Same visual language. No back-and-forth. Genspark AI Designer is not just a prompt box. This is scary good for solo builders and startups"  
[X Link](https://x.com/svpino/status/2019520437131337961)  2026-02-05T21:16Z 437.3K followers, [----] engagements


"Honestly managers firing their teams because rando42 said coding is dead kind of deserve whats coming to them. Darwinism at its best. I need folks to stop with the CODING IS DEAD talk because developers know that isn't the case but upper management thinks because these posts have a lot of likes they must be right. Tons of devs still work over [--] hours a week. I am very pro AI but this message isn't accurate. I need folks to stop with the CODING IS DEAD talk because developers know that isn't the case but upper management thinks because these posts have a lot of likes they must be right. Tons"  
[X Link](https://x.com/svpino/status/2019550138394624064)  2026-02-05T23:14Z 437.3K followers, 23.6K engagements


"I went from a 90/10 split between writing/reviewing code to around 20/80. That means I spend much more time looking at the code Claude writes than writing my own. Sometimes it's still easier to write Python than English but for the most part English is becoming my most-used programming language. I still don't understand the future implications of this. https://twitter.com/i/web/status/2019765543235170669 https://twitter.com/i/web/status/2019765543235170669"  
[X Link](https://x.com/svpino/status/2019765543235170669)  2026-02-06T13:30Z 437.5K followers, 19.9K engagements


"Prompt Engineering is dead. Context Engineering is king. Context is the biggest problem with coding agents. You can make any mediocre model work for you as long as you provide it with sufficient high-quality context"  
[X Link](https://x.com/svpino/status/2019851124354986294)  2026-02-06T19:10Z 437.6K followers, 31.7K engagements


"Here is how you can improve context using Claude Code Cursor Copilot Gemini CLI or any other agentic coding: You'll install @augmentcode's Context Engine. This is an MCP server that uses a knowledge graph and semantic indexing (not just keyword matching) to understand your code's structure and identify relevant code. When you are coding this MCP will give you: [--]. Better context [--]. Faster answers [--]. Lower token usage (cheaper) The larger your codebase the better (the more benefits you'll get from it). Here is how you can install it: [--]. Add the MCP server on Claude Code (link below) [--]. Update"  
[X Link](https://x.com/svpino/status/2019851137319596136)  2026-02-06T19:10Z 437.5K followers, [----] engagements


"I asked ChatGPT: "Draw a caricature of what you think of me based on what you know and the attached picture.""  
[X Link](https://x.com/svpino/status/2019882027554861295)  2026-02-06T21:12Z 437.3K followers, [----] engagements


"I write way less code now than I ever have. I also feel more in control of what I'm building than at any point in my career. Every line I do write carries more weight. Each unit of effort has an outsized influence on the final product. I'm not sure whether agentic coding is a net positive overall but so far it seems very helpful for my work. https://twitter.com/i/web/status/2020136472930840735 https://twitter.com/i/web/status/2020136472930840735"  
[X Link](https://x.com/svpino/status/2020136472930840735)  2026-02-07T14:04Z 437.5K followers, 13.9K engagements


"Gracias Benito Orgulloso de ser latino hoy ms que nunca"  
[X Link](https://x.com/svpino/status/2020704241372049594)  2026-02-09T03:40Z 437.4K followers, 12.1K engagements


"Genspark's also on the SuperBowl. AI is breaking out of its bubble and becoming a cultural phenomenon. The Monday after the Super Bowl should be a national holiday. We made a Super Bowl ad about it. Featuring Matthew Broderick and an AI that can actually autopilot your work. Airing today during the game ๐Ÿˆ https://t.co/BbZfDXVx5R The Monday after the Super Bowl should be a national holiday. We made a Super Bowl ad about it. Featuring Matthew Broderick and an AI that can actually autopilot your work. Airing today during the game ๐Ÿˆ https://t.co/BbZfDXVx5R"  
[X Link](https://x.com/svpino/status/2020856342014001277)  2026-02-09T13:44Z 437.4K followers, [----] engagements


"Agentic coding is not just about writing code anymore. They build the UI They send emails and handle notifications They set up authentication and user management They track analytics and monitor performance Mocha does this natively. Stop building dead products. The world doesn't need another "beautiful" app with zero users. If your platform can't talk to customers or track growth you're playing with templates. Today Mocha became the first AI Business Engine. Introducing Mocha Email & Mocha Analytics. https://t.co/P2cx2niLhv Stop building dead products. The world doesn't need another"  
[X Link](https://x.com/svpino/status/2020934266499236283)  2026-02-09T18:54Z 437.5K followers, [----] engagements


"What really excites me about Ai is that we aren't focusing on "writing code" anymore. That's table stakes. Now these agents: Build the UI Send emails and handle notifications Set up authentication and user management Track analytics and monitor performance It seems that every day we are automating more and more. [----] is off to a really great start https://twitter.com/i/web/status/2020934846197993960 https://twitter.com/i/web/status/2020934846197993960"  
[X Link](https://x.com/svpino/status/2020934846197993960)  2026-02-09T18:56Z 437.3K followers, [----] engagements


"You will never learn about AI agents by reading blog posts or watching tutorials. You can only learn by building. Here is a really cool AI challenge. (Great opportunity to learn and use different AI tools to solve a problem.) OutSystems is running its Groundhog Day Challenge. They are sponsoring this post. The challenge is a series of puzzles you can start solving without writing any code. As you make progress the challenges become more complex. They force you to build AI agents to crack them. This is a really good way to learn: Start simple Add complexity Go deeper and deeper These puzzles"  
[X Link](https://x.com/svpino/status/2021215176461254826)  2026-02-10T13:30Z 437.6K followers, 12.4K engagements


"Deploying what you build is what separates a real project from an academic exercise. I've already built a couple of dozen agents. Only three of them are in production. Building the agent is 20% of the work. Deploying it keeping it running handling failures scaling it and knowing when it breaks is the other 80%. Something I've learned the hard way: Deploying agents is 10x harder than deploying any other software I've built. Agents run continuously for minutes or hours Agents fail at unpredictable points Agents need to retry intelligently not blindly Most teams I talk to are duct-taping this"  
[X Link](https://x.com/svpino/status/2021255413270970546)  2026-02-10T16:10Z 437.6K followers, [----] engagements


"If you are a TypeScript developer check out Trigger. dev. They are collaborating with me on this post. It's an open-source platform for building and deploying AI agents and workflows. Here is what makes it interesting for anyone running agents in production: You can run agents for as long as they need to. No timeouts. Automatic retries with fine-grained control. You can configure retry logic per task or even per error type. Built-in queuing and concurrency controls. You decide how many agents run in parallel. Full observability. Trace every step of every run. You only pay when the code is"  
[X Link](https://x.com/svpino/status/2021255425778385390)  2026-02-10T16:10Z 437.5K followers, [----] engagements


"Here is a serious competitor for Lovable and Cursor. You can use it to build literally any app on any stack. Bring your own API key to use any model (GPT Claude Gemini etc.) Here is a cool Mario game built using it. Introducing Orchids [---] - the first AI app builder to build and deploy any app any stack (web mobile chrome extension slack bot AI agent anything). Use your ChatGPT Claude Code Github Copilot Gemini subscription - or any API key to use models at cost. Comment below to https://t.co/KmmZMXVGu8 Introducing Orchids [---] - the first AI app builder to build and deploy any app any stack"  
[X Link](https://x.com/svpino/status/2021260110392328267)  2026-02-10T16:28Z 437.5K followers, 11.9K engagements


"This is a serious competitor to Lovable and Cursor: is an AI-powered app builder: Build any app using any stack Bring your own API Key to use any model Integrates with any database payments API Full flexibility to build whatever you want: web applications mobile Chrome extensions Slack bots AI agents etc. Some really cool features: [--]. They have a "Record and Listen" feature where the AI can see your screen and hear your voice. It works like pair programming with a person. [--]. You can click on any UI element and describe what you want changed directly. No need to find it in code. [--]. Native"  
[X Link](https://x.com/svpino/status/2021315810715893870)  2026-02-10T20:10Z 437.6K followers, 11.6K engagements


"I'm glad we are already thinking about what comes after the IDE. Let's be honest: we've been using IDEs for a loooong time now. It's about time we find something better. What comes after the IDE This is our take. https://t.co/59EqpyJcs6 What comes after the IDE This is our take. https://t.co/59EqpyJcs6"  
[X Link](https://x.com/svpino/status/2021718313316036854)  2026-02-11T22:49Z 437.7K followers, 17.9K engagements


"When was the last time you read a good post-mortem I consulted for a company that famously joked about storing passwords in post-mortems because they knew no one would ever read them. This is the circle of life: Issue Fix Post-mortem Nobody reads it Same issue happens again a week later. Let's be honest: 99% of companies treat post-mortems as a compliance exercise. Meanwhile they are missing a huge opportunity to teach employees. The difference between a useless post-mortem and a good one is STRUCTURE. https://twitter.com/i/web/status/2021980137089368449"  
[X Link](https://x.com/svpino/status/2021980137089368449)  2026-02-12T16:10Z 437.7K followers, [----] engagements


"@pamelafox My respects to your mom :) I've spent hours teaching my dad how to hang up the phone and he still doesn't do it well"  
[X Link](https://x.com/svpino/status/2022042728314650784)  2026-02-12T20:18Z 437.6K followers, 11.3K engagements


"Claude Code tips that have made my life easier: (Add these to your CLAUDE .md file) [--]. "Before writing any code describe your approach and wait for approval. Always ask clarifying questions before writing any code if requirements are ambiguous." [--]. "If a task requires changes to more than [--] files stop and break it into smaller tasks first." [--]. "After writing code list what could break and suggest tests to cover it." [--]. "When theres a bug start by writing a test that reproduces it then fix it until the test passes." [--]. "Every time I correct you add a new rule to the CLAUDE .md file so it never"  
[X Link](https://x.com/svpino/status/2018682144361734368)  2026-02-03T13:45Z 437.6K followers, 224.1K engagements


"If you are using OpenClaw do this before it's too late: Scan every one of your skills using this Skill Scanner. Do not install new skills without first doing this. This app shows you exactly what skills OpenClaw can access before it runs. Do not let an AI agent run without understanding what it can do. Remember these agents have access to your files your data and external tools and they don't ask for permission. Gen Digital (the Fortune [---] company behind Norton and Avast) is building the "Agent Trust Hub". This is a trust layer for autonomous AI agents. The OpenClaw Skill Scanner is the"  
[X Link](https://x.com/svpino/status/2021339653555749252)  2026-02-10T21:45Z 437.6K followers, 64.7K engagements


"I went from writing code to writing specs. That's most of what I'm doing now. From writing loops and conditions to describing what I want to build. Basically I'm now operating at a much higher level of abstraction. Much more focus on the WHAT to build instead of the HOW to build it. I still write a lot of code mostly things that I don't know how to describe properly or the model can't get quite right. I also spend a ton of time reviewing code. Way more than ever before. A lot has changed for me over the last year or so. https://twitter.com/i/web/status/2021938094065516787"  
[X Link](https://x.com/svpino/status/2021938094065516787)  2026-02-12T13:23Z 437.6K followers, 21.6K engagements


"Best way to learn: Look at how other people solve their problems. The constraints they faced The tradeoffs they made The things they did to turn a demo into a system Theory is nice and all but case studies are 100x better. Books are nice and all but experience is 100x better. https://twitter.com/i/web/status/2022297229558730962 https://twitter.com/i/web/status/2022297229558730962"  
[X Link](https://x.com/svpino/status/2022297229558730962)  2026-02-13T13:10Z 437.6K followers, [----] engagements


"You'll get: The ability to run parallel agents Each agent with an isolated state Headless mode for CI/CD pipelines ACP support to connect to Neovim Zed Emacs etc. The barrier to trying agentic coding with this is now literally zero. Here is the link: Open-source is moving fast and I love to see it https://cline.bot/cli https://cline.bot/cli"  
[X Link](https://x.com/svpino/status/2022372729807405286)  2026-02-13T18:10Z 437.6K followers, [----] engagements


"We will not accept money to influence the answers we give you Google [----]. We are starting to test ads in ChatGPT free and Go (new $8/month option) tiers. Here are our principles. Most importantly we will not accept money to influence the answer ChatGPT gives you and we keep your conversations private from advertisers. It is clear to us that a lot We are starting to test ads in ChatGPT free and Go (new $8/month option) tiers. Here are our principles. Most importantly we will not accept money to influence the answer ChatGPT gives you and we keep your conversations private from advertisers. It"  
[X Link](https://x.com/svpino/status/2012282586890989719)  2026-01-16T21:55Z 437.6K followers, 623.5K engagements


"Great list but you cant become a software engineer today without learning about: - Large language models - AI-assisted coding - Agentic coding - Context engineering - Evaluation and verification - HITL workflows - Observability for AI systems If you want to become a world-class software engineer (in 2026) read these [--] books: If you want to become a world-class software engineer (in 2026) read these [--] books:"  
[X Link](https://x.com/svpino/status/2021295590391300446)  2026-02-10T18:49Z 437.6K followers, 29.2K engagements


"Literally the first transcription model that you can steer with a prompt. Send audio + a prompt. The model does what you ask. Universal [--] Pro from @AssemblyAI is the world's first production model to support this. [----] is off to a great start"  
[X Link](https://x.com/svpino/status/2022332642797158730)  2026-02-13T15:30Z 437.6K followers, [----] engagements


"open-source giving big ai a run for its money agentic coding for free with minimax m2.5 and kimi k.25 for free literally zero barriers to try agentic coding without paying a single cent Introducing Cline CLI 2.0: An open-source AI coding agent that runs entirely in your terminal. Parallel agents headless CI/CD pipelines ACP support for any editor and a completely redesigned developer experience. Minimax M2.5 and Kimi K2.5 are free to use for a limited time. https://t.co/R5TBKC71DZ Introducing Cline CLI 2.0: An open-source AI coding agent that runs entirely in your terminal. Parallel agents"  
[X Link](https://x.com/svpino/status/2022359704002007104)  2026-02-13T17:18Z 437.6K followers, 149.1K engagements


"The Caribbean is #1 but then South Florida beaches are a close second"  
[X Link](https://x.com/svpino/status/2023062173065650524)  2026-02-15T15:49Z 437.6K followers, [----] engagements


"Are you taking full advantage of Python [--] Are you sure Here are [--] Python [--] features that will change the way you are writing code today. ๐Ÿงต๐Ÿ‘‡"  
[X Link](https://x.com/svpino/status/1308632185113579522)  2020-09-23T05:00Z 437.7K followers, 13K engagements


"AI will not replace you. A person using AI will"  
[X Link](https://x.com/svpino/status/1610984481342771200)  2023-01-05T13:00Z 437.6K followers, 5.3M engagements


"My entire Master's program is free online. Yet most people won't go through it because they won't get a certificate. If you want to take the same classes I took to get my Master's degree here they are: https://www.udacity.com/georgia-tech https://www.udacity.com/georgia-tech"  
[X Link](https://x.com/svpino/status/1614970769116889093)  2023-01-16T13:00Z 437.7K followers, 761.9K engagements


"This will change education forever: [--]. Copy a lesson and paste it on ChatGPT. [--]. Ask it to formulate questions about the lesson. [--]. Answer the questions and ask ChatGPT to evaluate your answer. Look at the attached conversation"  
[X Link](https://x.com/svpino/status/1650107281433063428)  2023-04-23T12:00Z 437.7K followers, 841.9K engagements


"We integrated ChatGPT with our robots. We had a ton of fun building this Read on for the details:"  
[X Link](https://x.com/svpino/status/1650832349008125952)  2023-04-25T12:01Z 437.7K followers, 1.3M engagements


"I just spoke to myself and it was weird. You can now clone your voice using a few seconds of audio give the robot personality and deploy a clone of yourself. Here is a step-by-step process on how to do this in less than two minutes:"  
[X Link](https://x.com/svpino/status/1671488252568834048)  2023-06-21T12:00Z 437.7K followers, 2.3M engagements


"Here is an open-source generative AI application. It uses AI to generate professional headshots in minutes. It's an excellent opportunity for developers who want to see an end-to-end generative AI app from the inside. It's free and it's all in GitHub:"  
[X Link](https://x.com/svpino/status/1711003548073504886)  2023-10-08T13:00Z 437.7K followers, 1.5M engagements


"Here are [--] algorithms every programmer should know: Let's start with my top favorite [--]. If nothing else you should read about these algorithms and have a good idea of how they work: [--]. Linear search to find an element in a list [--]. Binary search to find an element on a sorted list [--]. Bubble sort to sort a list [--]. Merge sort will also sort lists [--]. Quicksort to sort the list and do it fast [--]. Dijkstra to find the shortest path in a graph [--]. Breadth-first Search (BFS) for trees or graphs [--]. Depth-first search (DFS) for trees or graphs [--]. Huffman for doing data compression [--]. Anything related"  
[X Link](https://x.com/svpino/status/1717875757198786851)  2023-10-27T12:07Z 437.7K followers, 801.2K engagements


"The best keyboard money can buy: Logitech MX Keys. I bought mine more than [---] years ago. In that time I've tried a dozen other keyboards but nothing comes close. Apple's keyboards are too flat and mechanical keyboards have too much travel. This one is perfect"  
[X Link](https://x.com/svpino/status/1780295142831948129)  2024-04-16T18:00Z 437.7K followers, 1.7M engagements


"Apple will blow everyone out of the water"  
[X Link](https://x.com/svpino/status/1808556102575141141)  2024-07-03T17:39Z 437.7K followers, 49.6M engagements


"The single most undervalued fact of linear algebra: Matrices are graphs and graphs are matrices. Encoding matrices as graphs is a cheat code making complex behavior simple to study. Let me show you how"  
[X Link](https://x.com/svpino/status/1822966303642308903)  2024-08-12T12:00Z 437.7K followers, 1.7M engagements


"My new soon-to-be Linux laptop right before I start assembling it"  
[X Link](https://x.com/svpino/status/1841133768502878666)  2024-10-01T15:11Z 437.7K followers, 2M engagements


"3 days using my Framework laptop running Ubuntu. Main takeaway: The Macbook Pro is the best laptop that has been created and nothing else is remotely close to it"  
[X Link](https://x.com/svpino/status/1842218315705630866)  2024-10-04T15:00Z 437.7K followers, 1.2M engagements


"China: Publishes an *open* model. *Open*AI: Publishes closed models gets beaten by China cries foul and tells us we are dumb because we are willing to send our data to China. Meanwhile OpenAI's Privacy Policy: "We collect EVERYTHING we can get from you.""  
[X Link](https://x.com/svpino/status/1883993852232753389)  2025-01-27T21:41Z 437.7K followers, 381K engagements


"Many years later the Logitech MX Keys is the best keyboard ever. No contest"  
[X Link](https://x.com/svpino/status/1916095208271581658)  2025-04-26T11:41Z 437.7K followers, 787.1K engagements


"Knowledge graphs are infinitely better than vector search for building the memory of AI agents. With five lines of code you can build a knowledge graph with your data. When you see the results you'll never go back to vector-mediocrity-land. Here is a quick video:"  
[X Link](https://x.com/svpino/status/1930221872005931515)  2025-06-04T11:15Z 437.7K followers, 410.2K engagements


"I just met a person who can't tell Python from C++. He has never written a single line of code in his life yet he feels he can build anything he wants. He told me point-blank: "I challenge you to tell me something I can't build using AI." I tried to explain but I couldn't find the right words. The most fascinating aspect of vibe-coding is how it has convinced so many people to believe they are better and more capable than they really are"  
[X Link](https://x.com/svpino/status/1956403013154902176)  2025-08-15T17:10Z 437.7K followers, 2.6M engagements


"This is the best laptop in the world. Period. M4 [--] GB RAM 1TB SSD [--] hours of battery life [--] inches [---] lbs. $1699 with the student discount. I've used it for [--] months and I haven't missed a Pro not even once. This laptop has no competition"  
[X Link](https://x.com/svpino/status/1962119440671429056)  2025-08-31T11:45Z 437.7K followers, 1.7M engagements


"Just like ChatGPT killed Google. Just like GPT-5 killed software engineering. Just like deep learning killed classical ML. Just like long context killed RAG. Just like MCP killed APIs. Just like synthetic data killed real data. Just like laptops killed desktops. Just like tablets killed laptops. Just like Web apps killed native apps. openai just killed n8n https://t.co/qFBdhGkY2G openai just killed n8n https://t.co/qFBdhGkY2G"  
[X Link](https://x.com/svpino/status/1975300122612285586)  2025-10-06T20:40Z 437.7K followers, 1.6M engagements


"I've never seen the Python community embrace any tool faster than they did with uv. uv is likely the best Python tool of the last few years. If you aren't using it yet stop what you are doing and look into it. If you are already a user check out the attached cheatsheet"  
[X Link](https://x.com/svpino/status/1978116122890448899)  2025-10-14T15:10Z 437.7K followers, 359.8K engagements


"Code without automated tests is as good as broken code. Testing was already important when people who knew what they were doing were writing the code and cared about the craft. Now AI-generated code is taking over. If you take a minute and look at the monstrosity that's coming out of vibe-coded software you'll understand how bad things are getting. Learn how to test your code. Go into debt if you have to. https://twitter.com/i/web/status/1994768593272410254 https://twitter.com/i/web/status/1994768593272410254"  
[X Link](https://x.com/svpino/status/1994768593272410254)  2025-11-29T14:01Z 437.6K followers, 38.7K engagements


"Is it just me or has ChatGPT become dumber It might just be that I'm using it more and asking more complex questions but my feeling over the last couple of weeks hasn't been good. It has become more verbose repetitive and it constantly flip-flops to the point where I need to fact-check every single thing it says because I can't trust it. I'm considering moving to Gemini as my daily driver and see what's up. https://twitter.com/i/web/status/2002017013171155096 https://twitter.com/i/web/status/2002017013171155096"  
[X Link](https://x.com/svpino/status/2002017013171155096)  2025-12-19T14:03Z 437.6K followers, 64.1K engagements


"Software development will never be the same. I want you to watch this video: This is a spec-driven development environment. 100% of your time goes to writing specs and managing agents. 0% goes to writing the code"  
[X Link](https://x.com/svpino/status/2002107789888655655)  2025-12-19T20:04Z 437.6K followers, 123.3K engagements


"This is a billion-dollar idea This is an open-source tool that detects signs of burnout in employees before it becomes a problem. The project is called On-Call Health and here's how it works: It combines two types of signals: [--]. Objective data from tools like PagerDuty Slack GitHub Linear and Jira. Think of incident volume after-hours pages number of tasks etc. [--]. Self-reported check-ins. These are short surveys asking engineers how they actually feel. Based on both the application computes a burnout risk score. For example if you have someone on your team who's been troubleshooting multiple"  
[X Link](https://x.com/svpino/status/2021602651029524690)  2026-02-11T15:10Z 437.6K followers, 16.8K engagements


"Open source is the opposite of what every big AI lab is offering. I was using the Cline CLI for the first time and they offer the freedom you won't get with any of the proprietary agentic coding tools: It's open-source You aren't locked into a single model Bring your own API Key to use any model Over [--] million developers already use it Fastest-growing AI open-source project on GitHub Even better: Right now you can use two of the best open source models for free: MiniMax M2.5 and Kimi K2.5. To install it: npm install -g cline https://twitter.com/i/web/status/2022372717748765167"  
[X Link](https://x.com/svpino/status/2022372717748765167)  2026-02-13T18:10Z 437.6K followers, 28.6K engagements


"Here is a completely free Computer Science curriculum for those with the discipline to go through it independently. The curriculum follows the same degree requirements as an undergraduate computer science major and consists of courses from Harvard MIT and Princeton among other colleges. The curriculum has four major sections: [--]. Introduction: Students can review this section and decide whether this fits them. [--]. Core: This section corresponds to the first three college years. [--]. Advanced: This section corresponds to the final year of a Computer Science curriculum in college. [--]. Final Project:"  
[X Link](https://x.com/svpino/status/1675808678086098944)  2023-07-03T10:08Z 437.6K followers, 3.1M engagements


"This is the best laptop in the world. Period. M4 [--] GB RAM 1TB SSD [--] hours of battery life [--] inches [---] lbs. $1699 with the student discount. No competition"  
[X Link](https://x.com/svpino/status/1900896021217243415)  2025-03-15T13:05Z 437.6K followers, 14.2M engagements


"Here is how you can give Claude Code access to any data that exists online. It's an easy way to make it 10x more powerful than it already is. For example: Use Claude Code to find open LinkedIn jobs in your area tailor your resume to them and apply for them automatically"  
[X Link](https://x.com/svpino/status/2016906799790923971)  2026-01-29T16:10Z 437.6K followers, 178K engagements


"me every time claude code one-shots something i said it was impossible for AI to do"  
[X Link](https://x.com/svpino/status/2018828943810294182)  2026-02-03T23:28Z 437.6K followers, 114.2K engagements


"A 100% automated QA team that works 24/7: [--]. Write your test in plain English [--]. AI generates the test cases [--]. Web agents execute the tests in parallel [--]. Live browser preview with everything that happens Try this for free in your project"  
[X Link](https://x.com/svpino/status/2020514022710538712)  2026-02-08T15:04Z 437.6K followers, 53.1K engagements


"@warpdotdev released Oz and it might be the missing piece we all need: Oz is a cloud-based coding agent with built-in orchestration (the *cloud* part is important because you don't have to run it locally). You set up isolated environments point them at your repos and manage everything from a single panel. You can see everything running in a single place (instead of having to tab-switch between terminals): Local agents Cloud agents Scheduled agents Agents triggered from the API (Slack GitHub Actions etc.) You can control every agent from that dashboard. You can even fork cloud agents to run on"  
[X Link](https://x.com/svpino/status/2022043280205398300)  2026-02-12T20:21Z 437.6K followers, [----] engagements


"big ai: software engineering is dead writing code is dead ai will write better binaries don't learn to write code everyone can build anything also big ai: "more than two colors is too hard""  
[X Link](https://x.com/svpino/status/2022672436286865460)  2026-02-14T14:01Z 437.6K followers, 150.6K engagements


"your css intern is the only ai-proof role at your company big ai: software engineering is dead writing code is dead ai will write better binaries don't learn to write code everyone can build anything also big ai: "more than two colors is too hard" https://t.co/K7IVpKaDdt big ai: software engineering is dead writing code is dead ai will write better binaries don't learn to write code everyone can build anything also big ai: "more than two colors is too hard" https://t.co/K7IVpKaDdt"  
[X Link](https://x.com/svpino/status/2022675192833851881)  2026-02-14T14:12Z 437.6K followers, [----] engagements


"Speculative decoding is not just "throughput optimization". In production systems the real impact of speculative decoding is on tail latency. Speculative decoding pairs your main model with a smaller faster draft model. The draft model proposes tokens; the main model verifies them in parallel. When the draft model predicts well you can skip serial generation steps and reduce latency. But there's a problem: A generic draft model wasn't trained on your data so they make many mistakes and the main model will reject most of their predictions. At this point any potential latency benefits are gone."  
[X Link](https://x.com/svpino/status/2022696088449532235)  2026-02-14T15:35Z 437.6K followers, [----] engagements


"Ive been an iTerm2 user since forever but lately I keep hearing about Ghostty. Is it worth switching With Claude Code Im spending an obscene amount of time in the terminal so its worth considering any potential improvements"  
[X Link](https://x.com/svpino/status/2023031802819166469)  2026-02-15T13:49Z 437.6K followers, 73.4K engagements


"Oh no They are stealing the stolen data OpenAI has warned US lawmakers that its Chinese rival DeepSeek is using unfair and increasingly sophisticated methods to extract results from leading US AI models to train the next generation of its breakthrough R1 chatbot https://t.co/siq46jRyM1 OpenAI has warned US lawmakers that its Chinese rival DeepSeek is using unfair and increasingly sophisticated methods to extract results from leading US AI models to train the next generation of its breakthrough R1 chatbot https://t.co/siq46jRyM1"  
[X Link](https://x.com/svpino/status/2022859362574774282)  2026-02-15T02:23Z 437.6K followers, 41.3K engagements


""open"claw just became open as in "open"ai Peter Steinberger is joining OpenAI to drive the next generation of personal agents. He is a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people. We expect this will quickly become core to our Peter Steinberger is joining OpenAI to drive the next generation of personal agents. He is a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people. We expect this will quickly become core to our"  
[X Link](https://x.com/svpino/status/2023181252871020927)  2026-02-15T23:42Z 437.6K followers, [----] engagements

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

@svpino Avatar @svpino Santiago

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

Engagements: [-------] #

Engagements Line Chart

  • [--] Week [---------] -9.10%
  • [--] Month [---------] +8.90%
  • [--] Months [----------] -40%
  • [--] Year [----------] -52%

Mentions: [--] #

Mentions Line Chart

  • [--] Week [---] +27%
  • [--] Month [---] +55%
  • [--] Months [---] +41%
  • [--] Year [-----] -8.60%

Followers: [-------] #

Followers Line Chart

  • [--] Week [-------] +0.21%
  • [--] Month [-------] +0.78%
  • [--] Months [-------] +3.70%
  • [--] Year [-------] +11%

CreatorRank: [------] #

CreatorRank Line Chart

Social Influence

Social category influence travel destinations #1909 technology brands 30% social networks 6% finance 6% stocks 3% products 1% countries 1%

Social topic influence ai #1786, code #419, claude code #48, if you 10%, build #1015, this is 8%, model #1840, how to 7%, agentic #97, engineering #106

Top accounts mentioned or mentioned by @lfuckingg @grok @mingtakaivo @gaekingai @unclejai @ghidorah_x @amanrawatamg @tacodevs @nabbilkhan @wildpinesai @pamelafox @ejae_dev @degenduke0 @saen_dev @cliwant_patrick @vlelyavin @rahul18kaushik @8snotreal @yungenfant @joshua_m_storm

Top assets mentioned Alphabet Inc Class A (GOOGL)

Top Social Posts

Top posts by engagements in the last [--] hours

"Most companies right now: - No automated tests - No code review process - No CI/CD pipelines - Poor secret management - No dataset versioning - Production workflows run from spreadsheets - No rollback plans - No integration tests These aren't just some weird companies. They're everywhere The market for people who can fix these fundamentals is massive. https://twitter.com/i/web/status/2017591212342128958 https://twitter.com/i/web/status/2017591212342128958"
X Link 2026-01-31T13:30Z 437.6K followers, 273.2K engagements

"You can set up moltbot (now openclaw) on Emergent in under [--] minutes. You don't need to buy a Mac mini use a terminal or worry about configuration files. Here's what you can do: [--]. Go to Emergent's homepage [--]. Click on the MoltBot option [--]. A prompt will automatically show up [--]. Click next That's it. You can then connect your WhatsApp or Telegram account to start using the bot immediately. This turns [--] hours of installation and configuration into [--] minutes. But even more important: This will help you keep things secure and away from your computer. MoltBot is fun and powerful but I do not"
X Link 2026-01-31T20:22Z 437.6K followers, 83.1K engagements

"This is literally the fastest way to install OpenClaw (MoltBot). This video will show you how to do it step by step. You don't need to buy a Mac Mini. We'll install it in @digitalocean. I've installed this [--] different times already and this is the fastest way I've found"
X Link 2026-02-04T14:01Z 437.6K followers, 239K engagements

"Claude Code /insights command is insanely useful Do this: [--]. Run the /insights command [--]. Find the generated report file [--]. Attach it to a Claude session Prompt: Read the attached report and create a bullet list with [--] lessons to improve my use of Claude Code. Mind blown"
X Link 2026-02-11T13:34Z 437.6K followers, 34.8K engagements

"MiniMax M2.1 is really good for writing code. It's open source and you can connect Visual Studio Code to it and start coding right away. The model is especially good for those who don't use Python as a primary language. It's currently the best multilingual model out there. The model hits 72.5% on SWE-Multilingual and 88.6% on VIBE-Bench (best) This means the model is also really good at performing real engineering work and does very well at agentic tasks: Long tool chains Structured agent scaffolding Following repo-level rules and slash commands Working inside existing agent frameworks"
X Link 2026-01-26T20:01Z 437.6K followers, 48.3K engagements

"I'm giving a free talk at Sonar Summit in March. I'll focus on Claude Code and share dozens of tips for using it more effectively. My talk will be FREE and ONLINE. Everyone can attend. RSVP using this link: https://fandf.co/406EbDf https://fandf.co/406EbDf"
X Link 2026-02-11T13:34Z 437.6K followers, [----] engagements

"By the way the insights report that Claude Code generates is packed with useful information. You can simply read the report yourself and it will give you a bunch of ideas that you can implement to improve it. There's a lot of information so I like to use Claude to distill what's important for me. Some examples I got from my report (these might not make sense for everyone because they are specific to my use cases): [--]. The report recommended including my file structure in CLAUDE .md [--]. Name the tool in my prompts when finding business information [--]. Add a PreToolUse hook to run a function 4."
X Link 2026-02-11T13:34Z 437.6K followers, [----] engagements

"sOfTwArE eNgInEeRiNg iS dEaD you have to be mentally challenged to think your mom will use ai to build a grocery app whenever she needs one"
X Link 2026-02-12T17:16Z 437.6K followers, 304.4K engagements

"I've been running 4-6 coding agents every single day for months now: I have [--] - [--] instances of Claude Code. I have Copilot. I have Warp. I have Jules. I'm figuring this out as I go but it's getting really hard to keep track of everything that's happening. 90% of the time I can't remember which agent is working on what. I can't keep context in my head and I forget to check the agent that finished [--] minutes ago. Models aren't the bottleneck anymore. I am. I think there's one solution for this: https://twitter.com/i/web/status/2022043267945443337"
X Link 2026-02-12T20:20Z 437.6K followers, 20.4K engagements

"engineering team shipping dark mode during a DDoS attack BREAKING: House Republicans announce investigation into Bad Bunnys Super Bowl performance alleging the Spanish lyrics were inappropriate for television. BREAKING: House Republicans announce investigation into Bad Bunnys Super Bowl performance alleging the Spanish lyrics were inappropriate for television"
X Link 2026-02-13T15:29Z 437.6K followers, 11K engagements

"Honestly only nerds care about coding agents. What your mom friends and neighbors want is something that does real things for them. Lindy is a pretty dope assistant. Nothing to install and nothing to configure. You don't need a technical cousin to run this for you. Here is a quick list of what it does: It checks your email It can reply back using your voice It schedules meetings It pulls context from past conversations before meetings It takes notes during meetings It tells you what's important It can draft and send follow-ups It can text you on iMessage when it needs you. This is the type"
X Link 2026-02-13T20:55Z 437.6K followers, 15.1K engagements

"@BelieveOnJesus The fact that AI already replaced 100% of your coding tasks says a lot about the complexity of the work you were doing before"
X Link 2026-02-14T14:49Z 437.6K followers, [----] engagements

"@_flashman_harry Sure thing. We are also colonizing Mars. sorry I meant the Moon and we will have 100% self-driving cars by the end of 2020"
X Link 2026-02-14T15:13Z 437.6K followers, [----] engagements

"You are so close of getting it Its so hard to contrast Boris take with Darios claim that: software engineering is fully automated by [----]. You might have the 0.01% engineer left over but that's it the job is finished. Why does Dario keep making such insane claims Its so hard to contrast Boris take with Darios claim that: software engineering is fully automated by [----]. You might have the 0.01% engineer left over but that's it the job is finished. Why does Dario keep making such insane claims"
X Link 2026-02-15T03:27Z 437.6K followers, 27.7K engagements

"11 ways ChatGPT saves me hours of work every day and why you'll never outcompete those who use AI effectively. A list for those who write code: [--] of 16"
X Link 2023-02-11T13:00Z 437.7K followers, 2.1M engagements

"Another deep learning breakthrough: Deep TDA a new algorithm using self-supervised learning overcomes the limitations of traditional dimensionality reduction algorithms. t-SNE and UMAP have long been the favorites. Deep TDA might change that forever. Here are the details:"
X Link 2023-07-05T12:00Z 436.9K followers, 1.1M engagements

"Best GPU prices you'll find out there: H100 GPU for $2.00/hr H200 GPU for $2.30/hr Even better: You can literally move all of your data to Nebius for free including a free PoC with support and get up to [--] months for free Nebius is collaborating with me on this post and their deal is just amazing This is insanely cheap and you should check it out if you are looking to fine-tune models run experiments or do some production testing. Check this out: https://bit.ly/46rANG7 https://bit.ly/46rANG7"
X Link 2025-09-22T15:10Z 436.9K followers, 59.3K engagements

"I still think that prompt engineering is not a career. I still think companies will not start hiring prompt engineers anytime soon. I still think about prompt engineering as a skill not a role. Here is a counterintuitive lesson from building agents: The prompt is often more important than the system's architecture. I've always been obsessed with breaking down my system into the proper number of agents and assigning responsibilities correctly but the biggest gains in Here is a counterintuitive lesson from building agents: The prompt is often more important than the system's architecture. I've"
X Link 2025-09-26T12:55Z 437K followers, 15.8K engagements

"Building software is only 10% about writing code and 90% about thinking what code to write why you should write it and how to do it well. You can delegate the code-writing part to an LLM but so far they have shown they cant handle the thinking part. Its called vibe-coding and not vibe-thinking for a reason. Vibe-coders cant build the type of software $400000/year salaries pay today. If you believe companies pay that money to build vibe-code-able apps you should crack a book and get off Twitter. Vibe-coders cant build the type of software $400000/year salaries pay today. If you believe"
X Link 2025-10-12T15:31Z 437.6K followers, 168.1K engagements

"Ive been building software for [--] years more than [--] professionally. I understand architecture principles and best practices. No you cant vibe-code Salesforce SAP or any other complex enterprise app. @svpino @Rifadm816 If you understand architecture principles and best practices then yes absolutely. @svpino @Rifadm816 If you understand architecture principles and best practices then yes absolutely"
X Link 2025-11-26T19:22Z 437.6K followers, 374.2K engagements

"I don't believe you can vibe-code complex applications. Put another way your ability to vibe-code software is inversely proportional to its complexity. Vibe-coding is like going into the "Minotaur Labyrinth" without Theseus' thread: you'll invariably get lost. More accurately you'll be stuck in an endless loop going in circles with the model: Things that used to work stop working Your code starts bloating Solutions become over-engineered And all of this will happen while working with popular tech where models have plenty of training. Try to step outside the beaten path (use anything a bit"
X Link 2025-11-27T13:10Z 437.7K followers, 78.5K engagements

"There is a massive opportunity right now for anyone who can walk into a company and automate the grunt work they do every day. It doesn't matter what programming language you use or libraries you like or whether you know what micro-services do or if you like Claude Code or Cursor. No one cares about any of that. If you can identify a problem and convince them you can solve it you'll make a killing"
X Link 2025-11-28T13:45Z 436.9K followers, 50.2K engagements

"68% of developers say they spend more time debugging AI code than writing new code. (According to a survey of [---] software engineering leaders and practitioners.) AI coding assistants help us write code faster but they increase the amount of insecure and broken code we ship"
X Link 2026-01-19T15:01Z 437.6K followers, 36.9K engagements

"All these [--] models are now available: [--]. Qwen3-Next-80B-A3B-Thinking [--]. DeepSeek-V3.2 [--]. MiniMax M2.1 [--]. Nemotron-3-Nano-30B-A3B Open source AI is unstoppable. Excited to announce two new models added to Token Factory First: @nvidia Nemotron-3-Nano-30B-A3B a compact MoE (30B params 3.6B active) for coding reasoning instruction-following tool-calling. Second: @Alibaba_Qwen Qwen3-Next-80B-A3B-Thinking a powerful MoE (80B params https://t.co/EIlt53FhHK Excited to announce two new models added to Token Factory First: @nvidia Nemotron-3-Nano-30B-A3B a compact MoE (30B params 3.6B active) for"
X Link 2026-01-23T21:02Z 437.3K followers, 31.1K engagements

"Voice agents are awkward and everyone notices: You ask a question. The agent thinks. You wait. And wait. Nobody wants this. I'd rather talk to a person. If your model's response time is over 300ms you won't make it. Unfortunately most text-to-speech models can't get anywhere close to that. I want you to take a look at the latest model released by @inworld_ai: TTS-1.5. I built a simple voice agent using the model so you can see it in action and test it on your computer. You'll find the repository link below. The latency numbers of this model are wild: Max model under 250ms Mini model under"
X Link 2026-01-27T16:10Z 437.7K followers, 69.2K engagements

"GitHub just released its Copilot SDK. You can now embed Copilot into any of your applications and make it much more powerful with just a few lines of code. I built a quick example here"
X Link 2026-01-28T16:01Z 437.5K followers, 60.7K engagements

"I don't remember the last time I was so excited about something as I am right now about Claude Code"
X Link 2026-02-02T13:30Z 437.3K followers, 51.1K engagements

"I've been trying to move from imperative prompting to declarative prompting. Imperative prompting: I tell the agent how to do something. Declarative prompting: I describe my goal to the agent. This was impossible before but since Opus [---] I'm letting models fill in the blanks more and more. I recorded a video using Quest + Opus [---]. This is a coding agent designed specifically for this purpose"
X Link 2026-02-02T16:10Z 436.9K followers, 96.6K engagements

"So many people are shipping AI-generated code without even looking at it. It's terrifying AI models are improving but they still make a ton of mistakes. Now more than ever you get code that looks fine and passes a quick glance but underneath it doesn't work: race conditions broken contracts duplicated logic and any number of subtle issues that aren't obvious. This will come back to haunt us forever. The team behind Qodo just launched their [---] version to address this. https://twitter.com/i/web/status/2019140163499352527 https://twitter.com/i/web/status/2019140163499352527"
X Link 2026-02-04T20:05Z 437K followers, 19.1K engagements

"Sam meet Sam. First the good part of the Anthropic ads: they are funny and I laughed. But I wonder why Anthropic would go for something so clearly dishonest. Our most important principle for ads says that we wont do exactly this; we would obviously never run ads in the way Anthropic First the good part of the Anthropic ads: they are funny and I laughed. But I wonder why Anthropic would go for something so clearly dishonest. Our most important principle for ads says that we wont do exactly this; we would obviously never run ads in the way Anthropic"
X Link 2026-02-04T20:48Z 437.5K followers, 96K engagements

"There are so many people telling these amazing stories of how they built these great applications overnight using AI and how they are making so much money because of it And I think its all bullshit. Ive been using these models every day since they were 10x more dumb they are today and theres absolutely no chance non-devs are one-shotting anything of value without significant manual work. My honest opinion: people who claim to have easily "one shotted" something with AI probably work on very conventional use cases. Or have ready made templates of steps that need to be accomplished. Every"
X Link 2026-02-05T00:02Z 437.6K followers, 94.4K engagements

"Opus [---] is so good that declarative prompting is now becoming possible for the first time. Here is how I usually think about prompting the model: "Rewrite this function using a hash map instead of nested loops." That's "imperative prompting." With the latest iteration you can now try this instead: "This function is too slow. Make it faster without changing what it returns." That's declarative prompting. This doesn't always work (the agent sometimes needs more context and details to not screw things up) but my success rate is much higher than ever before."
X Link 2026-02-05T14:01Z 437.6K followers, 71.9K engagements

"Testing sucks across most companies I work with. I see this over and over again: Most teams don't write automated tests. Some of them do some formal manual testing. Some do have automated tests but nobody trusts them. Writing automated tests is hard. Maintaining them is even harder. Every UI change every new feature every refactor breaks something and someone has to go fix the tests. AI can help here. Look at the attached video"
X Link 2026-02-05T18:07Z 437.3K followers, 13.4K engagements

"Virtual assistants should be scared @genspark_ai just hit $155M in ARR in only [--] months. This is an all-in-one AI workspace that replaces multiple tools: Slides Design Images Data Research And it's all inside one interface. Here's the crazy part: It's $19.99/month and you get access to the best AI models in the world + a specialized agents that do the work for you"
X Link 2026-02-05T21:15Z 437.3K followers, [----] engagements

"You can also ask Genspark to build a website similar to another. Same style. Same visual language. No back-and-forth. Genspark AI Designer is not just a prompt box. This is scary good for solo builders and startups"
X Link 2026-02-05T21:16Z 437.3K followers, [----] engagements

"Honestly managers firing their teams because rando42 said coding is dead kind of deserve whats coming to them. Darwinism at its best. I need folks to stop with the CODING IS DEAD talk because developers know that isn't the case but upper management thinks because these posts have a lot of likes they must be right. Tons of devs still work over [--] hours a week. I am very pro AI but this message isn't accurate. I need folks to stop with the CODING IS DEAD talk because developers know that isn't the case but upper management thinks because these posts have a lot of likes they must be right. Tons"
X Link 2026-02-05T23:14Z 437.3K followers, 23.6K engagements

"I went from a 90/10 split between writing/reviewing code to around 20/80. That means I spend much more time looking at the code Claude writes than writing my own. Sometimes it's still easier to write Python than English but for the most part English is becoming my most-used programming language. I still don't understand the future implications of this. https://twitter.com/i/web/status/2019765543235170669 https://twitter.com/i/web/status/2019765543235170669"
X Link 2026-02-06T13:30Z 437.5K followers, 19.9K engagements

"Prompt Engineering is dead. Context Engineering is king. Context is the biggest problem with coding agents. You can make any mediocre model work for you as long as you provide it with sufficient high-quality context"
X Link 2026-02-06T19:10Z 437.6K followers, 31.7K engagements

"Here is how you can improve context using Claude Code Cursor Copilot Gemini CLI or any other agentic coding: You'll install @augmentcode's Context Engine. This is an MCP server that uses a knowledge graph and semantic indexing (not just keyword matching) to understand your code's structure and identify relevant code. When you are coding this MCP will give you: [--]. Better context [--]. Faster answers [--]. Lower token usage (cheaper) The larger your codebase the better (the more benefits you'll get from it). Here is how you can install it: [--]. Add the MCP server on Claude Code (link below) [--]. Update"
X Link 2026-02-06T19:10Z 437.5K followers, [----] engagements

"I asked ChatGPT: "Draw a caricature of what you think of me based on what you know and the attached picture.""
X Link 2026-02-06T21:12Z 437.3K followers, [----] engagements

"I write way less code now than I ever have. I also feel more in control of what I'm building than at any point in my career. Every line I do write carries more weight. Each unit of effort has an outsized influence on the final product. I'm not sure whether agentic coding is a net positive overall but so far it seems very helpful for my work. https://twitter.com/i/web/status/2020136472930840735 https://twitter.com/i/web/status/2020136472930840735"
X Link 2026-02-07T14:04Z 437.5K followers, 13.9K engagements

"Gracias Benito Orgulloso de ser latino hoy ms que nunca"
X Link 2026-02-09T03:40Z 437.4K followers, 12.1K engagements

"Genspark's also on the SuperBowl. AI is breaking out of its bubble and becoming a cultural phenomenon. The Monday after the Super Bowl should be a national holiday. We made a Super Bowl ad about it. Featuring Matthew Broderick and an AI that can actually autopilot your work. Airing today during the game ๐Ÿˆ https://t.co/BbZfDXVx5R The Monday after the Super Bowl should be a national holiday. We made a Super Bowl ad about it. Featuring Matthew Broderick and an AI that can actually autopilot your work. Airing today during the game ๐Ÿˆ https://t.co/BbZfDXVx5R"
X Link 2026-02-09T13:44Z 437.4K followers, [----] engagements

"Agentic coding is not just about writing code anymore. They build the UI They send emails and handle notifications They set up authentication and user management They track analytics and monitor performance Mocha does this natively. Stop building dead products. The world doesn't need another "beautiful" app with zero users. If your platform can't talk to customers or track growth you're playing with templates. Today Mocha became the first AI Business Engine. Introducing Mocha Email & Mocha Analytics. https://t.co/P2cx2niLhv Stop building dead products. The world doesn't need another"
X Link 2026-02-09T18:54Z 437.5K followers, [----] engagements

"What really excites me about Ai is that we aren't focusing on "writing code" anymore. That's table stakes. Now these agents: Build the UI Send emails and handle notifications Set up authentication and user management Track analytics and monitor performance It seems that every day we are automating more and more. [----] is off to a really great start https://twitter.com/i/web/status/2020934846197993960 https://twitter.com/i/web/status/2020934846197993960"
X Link 2026-02-09T18:56Z 437.3K followers, [----] engagements

"You will never learn about AI agents by reading blog posts or watching tutorials. You can only learn by building. Here is a really cool AI challenge. (Great opportunity to learn and use different AI tools to solve a problem.) OutSystems is running its Groundhog Day Challenge. They are sponsoring this post. The challenge is a series of puzzles you can start solving without writing any code. As you make progress the challenges become more complex. They force you to build AI agents to crack them. This is a really good way to learn: Start simple Add complexity Go deeper and deeper These puzzles"
X Link 2026-02-10T13:30Z 437.6K followers, 12.4K engagements

"Deploying what you build is what separates a real project from an academic exercise. I've already built a couple of dozen agents. Only three of them are in production. Building the agent is 20% of the work. Deploying it keeping it running handling failures scaling it and knowing when it breaks is the other 80%. Something I've learned the hard way: Deploying agents is 10x harder than deploying any other software I've built. Agents run continuously for minutes or hours Agents fail at unpredictable points Agents need to retry intelligently not blindly Most teams I talk to are duct-taping this"
X Link 2026-02-10T16:10Z 437.6K followers, [----] engagements

"If you are a TypeScript developer check out Trigger. dev. They are collaborating with me on this post. It's an open-source platform for building and deploying AI agents and workflows. Here is what makes it interesting for anyone running agents in production: You can run agents for as long as they need to. No timeouts. Automatic retries with fine-grained control. You can configure retry logic per task or even per error type. Built-in queuing and concurrency controls. You decide how many agents run in parallel. Full observability. Trace every step of every run. You only pay when the code is"
X Link 2026-02-10T16:10Z 437.5K followers, [----] engagements

"Here is a serious competitor for Lovable and Cursor. You can use it to build literally any app on any stack. Bring your own API key to use any model (GPT Claude Gemini etc.) Here is a cool Mario game built using it. Introducing Orchids [---] - the first AI app builder to build and deploy any app any stack (web mobile chrome extension slack bot AI agent anything). Use your ChatGPT Claude Code Github Copilot Gemini subscription - or any API key to use models at cost. Comment below to https://t.co/KmmZMXVGu8 Introducing Orchids [---] - the first AI app builder to build and deploy any app any stack"
X Link 2026-02-10T16:28Z 437.5K followers, 11.9K engagements

"This is a serious competitor to Lovable and Cursor: is an AI-powered app builder: Build any app using any stack Bring your own API Key to use any model Integrates with any database payments API Full flexibility to build whatever you want: web applications mobile Chrome extensions Slack bots AI agents etc. Some really cool features: [--]. They have a "Record and Listen" feature where the AI can see your screen and hear your voice. It works like pair programming with a person. [--]. You can click on any UI element and describe what you want changed directly. No need to find it in code. [--]. Native"
X Link 2026-02-10T20:10Z 437.6K followers, 11.6K engagements

"I'm glad we are already thinking about what comes after the IDE. Let's be honest: we've been using IDEs for a loooong time now. It's about time we find something better. What comes after the IDE This is our take. https://t.co/59EqpyJcs6 What comes after the IDE This is our take. https://t.co/59EqpyJcs6"
X Link 2026-02-11T22:49Z 437.7K followers, 17.9K engagements

"When was the last time you read a good post-mortem I consulted for a company that famously joked about storing passwords in post-mortems because they knew no one would ever read them. This is the circle of life: Issue Fix Post-mortem Nobody reads it Same issue happens again a week later. Let's be honest: 99% of companies treat post-mortems as a compliance exercise. Meanwhile they are missing a huge opportunity to teach employees. The difference between a useless post-mortem and a good one is STRUCTURE. https://twitter.com/i/web/status/2021980137089368449"
X Link 2026-02-12T16:10Z 437.7K followers, [----] engagements

"@pamelafox My respects to your mom :) I've spent hours teaching my dad how to hang up the phone and he still doesn't do it well"
X Link 2026-02-12T20:18Z 437.6K followers, 11.3K engagements

"Claude Code tips that have made my life easier: (Add these to your CLAUDE .md file) [--]. "Before writing any code describe your approach and wait for approval. Always ask clarifying questions before writing any code if requirements are ambiguous." [--]. "If a task requires changes to more than [--] files stop and break it into smaller tasks first." [--]. "After writing code list what could break and suggest tests to cover it." [--]. "When theres a bug start by writing a test that reproduces it then fix it until the test passes." [--]. "Every time I correct you add a new rule to the CLAUDE .md file so it never"
X Link 2026-02-03T13:45Z 437.6K followers, 224.1K engagements

"If you are using OpenClaw do this before it's too late: Scan every one of your skills using this Skill Scanner. Do not install new skills without first doing this. This app shows you exactly what skills OpenClaw can access before it runs. Do not let an AI agent run without understanding what it can do. Remember these agents have access to your files your data and external tools and they don't ask for permission. Gen Digital (the Fortune [---] company behind Norton and Avast) is building the "Agent Trust Hub". This is a trust layer for autonomous AI agents. The OpenClaw Skill Scanner is the"
X Link 2026-02-10T21:45Z 437.6K followers, 64.7K engagements

"I went from writing code to writing specs. That's most of what I'm doing now. From writing loops and conditions to describing what I want to build. Basically I'm now operating at a much higher level of abstraction. Much more focus on the WHAT to build instead of the HOW to build it. I still write a lot of code mostly things that I don't know how to describe properly or the model can't get quite right. I also spend a ton of time reviewing code. Way more than ever before. A lot has changed for me over the last year or so. https://twitter.com/i/web/status/2021938094065516787"
X Link 2026-02-12T13:23Z 437.6K followers, 21.6K engagements

"Best way to learn: Look at how other people solve their problems. The constraints they faced The tradeoffs they made The things they did to turn a demo into a system Theory is nice and all but case studies are 100x better. Books are nice and all but experience is 100x better. https://twitter.com/i/web/status/2022297229558730962 https://twitter.com/i/web/status/2022297229558730962"
X Link 2026-02-13T13:10Z 437.6K followers, [----] engagements

"You'll get: The ability to run parallel agents Each agent with an isolated state Headless mode for CI/CD pipelines ACP support to connect to Neovim Zed Emacs etc. The barrier to trying agentic coding with this is now literally zero. Here is the link: Open-source is moving fast and I love to see it https://cline.bot/cli https://cline.bot/cli"
X Link 2026-02-13T18:10Z 437.6K followers, [----] engagements

"We will not accept money to influence the answers we give you Google [----]. We are starting to test ads in ChatGPT free and Go (new $8/month option) tiers. Here are our principles. Most importantly we will not accept money to influence the answer ChatGPT gives you and we keep your conversations private from advertisers. It is clear to us that a lot We are starting to test ads in ChatGPT free and Go (new $8/month option) tiers. Here are our principles. Most importantly we will not accept money to influence the answer ChatGPT gives you and we keep your conversations private from advertisers. It"
X Link 2026-01-16T21:55Z 437.6K followers, 623.5K engagements

"Great list but you cant become a software engineer today without learning about: - Large language models - AI-assisted coding - Agentic coding - Context engineering - Evaluation and verification - HITL workflows - Observability for AI systems If you want to become a world-class software engineer (in 2026) read these [--] books: If you want to become a world-class software engineer (in 2026) read these [--] books:"
X Link 2026-02-10T18:49Z 437.6K followers, 29.2K engagements

"Literally the first transcription model that you can steer with a prompt. Send audio + a prompt. The model does what you ask. Universal [--] Pro from @AssemblyAI is the world's first production model to support this. [----] is off to a great start"
X Link 2026-02-13T15:30Z 437.6K followers, [----] engagements

"open-source giving big ai a run for its money agentic coding for free with minimax m2.5 and kimi k.25 for free literally zero barriers to try agentic coding without paying a single cent Introducing Cline CLI 2.0: An open-source AI coding agent that runs entirely in your terminal. Parallel agents headless CI/CD pipelines ACP support for any editor and a completely redesigned developer experience. Minimax M2.5 and Kimi K2.5 are free to use for a limited time. https://t.co/R5TBKC71DZ Introducing Cline CLI 2.0: An open-source AI coding agent that runs entirely in your terminal. Parallel agents"
X Link 2026-02-13T17:18Z 437.6K followers, 149.1K engagements

"The Caribbean is #1 but then South Florida beaches are a close second"
X Link 2026-02-15T15:49Z 437.6K followers, [----] engagements

"Are you taking full advantage of Python [--] Are you sure Here are [--] Python [--] features that will change the way you are writing code today. ๐Ÿงต๐Ÿ‘‡"
X Link 2020-09-23T05:00Z 437.7K followers, 13K engagements

"AI will not replace you. A person using AI will"
X Link 2023-01-05T13:00Z 437.6K followers, 5.3M engagements

"My entire Master's program is free online. Yet most people won't go through it because they won't get a certificate. If you want to take the same classes I took to get my Master's degree here they are: https://www.udacity.com/georgia-tech https://www.udacity.com/georgia-tech"
X Link 2023-01-16T13:00Z 437.7K followers, 761.9K engagements

"This will change education forever: [--]. Copy a lesson and paste it on ChatGPT. [--]. Ask it to formulate questions about the lesson. [--]. Answer the questions and ask ChatGPT to evaluate your answer. Look at the attached conversation"
X Link 2023-04-23T12:00Z 437.7K followers, 841.9K engagements

"We integrated ChatGPT with our robots. We had a ton of fun building this Read on for the details:"
X Link 2023-04-25T12:01Z 437.7K followers, 1.3M engagements

"I just spoke to myself and it was weird. You can now clone your voice using a few seconds of audio give the robot personality and deploy a clone of yourself. Here is a step-by-step process on how to do this in less than two minutes:"
X Link 2023-06-21T12:00Z 437.7K followers, 2.3M engagements

"Here is an open-source generative AI application. It uses AI to generate professional headshots in minutes. It's an excellent opportunity for developers who want to see an end-to-end generative AI app from the inside. It's free and it's all in GitHub:"
X Link 2023-10-08T13:00Z 437.7K followers, 1.5M engagements

"Here are [--] algorithms every programmer should know: Let's start with my top favorite [--]. If nothing else you should read about these algorithms and have a good idea of how they work: [--]. Linear search to find an element in a list [--]. Binary search to find an element on a sorted list [--]. Bubble sort to sort a list [--]. Merge sort will also sort lists [--]. Quicksort to sort the list and do it fast [--]. Dijkstra to find the shortest path in a graph [--]. Breadth-first Search (BFS) for trees or graphs [--]. Depth-first search (DFS) for trees or graphs [--]. Huffman for doing data compression [--]. Anything related"
X Link 2023-10-27T12:07Z 437.7K followers, 801.2K engagements

"The best keyboard money can buy: Logitech MX Keys. I bought mine more than [---] years ago. In that time I've tried a dozen other keyboards but nothing comes close. Apple's keyboards are too flat and mechanical keyboards have too much travel. This one is perfect"
X Link 2024-04-16T18:00Z 437.7K followers, 1.7M engagements

"Apple will blow everyone out of the water"
X Link 2024-07-03T17:39Z 437.7K followers, 49.6M engagements

"The single most undervalued fact of linear algebra: Matrices are graphs and graphs are matrices. Encoding matrices as graphs is a cheat code making complex behavior simple to study. Let me show you how"
X Link 2024-08-12T12:00Z 437.7K followers, 1.7M engagements

"My new soon-to-be Linux laptop right before I start assembling it"
X Link 2024-10-01T15:11Z 437.7K followers, 2M engagements

"3 days using my Framework laptop running Ubuntu. Main takeaway: The Macbook Pro is the best laptop that has been created and nothing else is remotely close to it"
X Link 2024-10-04T15:00Z 437.7K followers, 1.2M engagements

"China: Publishes an open model. OpenAI: Publishes closed models gets beaten by China cries foul and tells us we are dumb because we are willing to send our data to China. Meanwhile OpenAI's Privacy Policy: "We collect EVERYTHING we can get from you.""
X Link 2025-01-27T21:41Z 437.7K followers, 381K engagements

"Many years later the Logitech MX Keys is the best keyboard ever. No contest"
X Link 2025-04-26T11:41Z 437.7K followers, 787.1K engagements

"Knowledge graphs are infinitely better than vector search for building the memory of AI agents. With five lines of code you can build a knowledge graph with your data. When you see the results you'll never go back to vector-mediocrity-land. Here is a quick video:"
X Link 2025-06-04T11:15Z 437.7K followers, 410.2K engagements

"I just met a person who can't tell Python from C++. He has never written a single line of code in his life yet he feels he can build anything he wants. He told me point-blank: "I challenge you to tell me something I can't build using AI." I tried to explain but I couldn't find the right words. The most fascinating aspect of vibe-coding is how it has convinced so many people to believe they are better and more capable than they really are"
X Link 2025-08-15T17:10Z 437.7K followers, 2.6M engagements

"This is the best laptop in the world. Period. M4 [--] GB RAM 1TB SSD [--] hours of battery life [--] inches [---] lbs. $1699 with the student discount. I've used it for [--] months and I haven't missed a Pro not even once. This laptop has no competition"
X Link 2025-08-31T11:45Z 437.7K followers, 1.7M engagements

"Just like ChatGPT killed Google. Just like GPT-5 killed software engineering. Just like deep learning killed classical ML. Just like long context killed RAG. Just like MCP killed APIs. Just like synthetic data killed real data. Just like laptops killed desktops. Just like tablets killed laptops. Just like Web apps killed native apps. openai just killed n8n https://t.co/qFBdhGkY2G openai just killed n8n https://t.co/qFBdhGkY2G"
X Link 2025-10-06T20:40Z 437.7K followers, 1.6M engagements

"I've never seen the Python community embrace any tool faster than they did with uv. uv is likely the best Python tool of the last few years. If you aren't using it yet stop what you are doing and look into it. If you are already a user check out the attached cheatsheet"
X Link 2025-10-14T15:10Z 437.7K followers, 359.8K engagements

"Code without automated tests is as good as broken code. Testing was already important when people who knew what they were doing were writing the code and cared about the craft. Now AI-generated code is taking over. If you take a minute and look at the monstrosity that's coming out of vibe-coded software you'll understand how bad things are getting. Learn how to test your code. Go into debt if you have to. https://twitter.com/i/web/status/1994768593272410254 https://twitter.com/i/web/status/1994768593272410254"
X Link 2025-11-29T14:01Z 437.6K followers, 38.7K engagements

"Is it just me or has ChatGPT become dumber It might just be that I'm using it more and asking more complex questions but my feeling over the last couple of weeks hasn't been good. It has become more verbose repetitive and it constantly flip-flops to the point where I need to fact-check every single thing it says because I can't trust it. I'm considering moving to Gemini as my daily driver and see what's up. https://twitter.com/i/web/status/2002017013171155096 https://twitter.com/i/web/status/2002017013171155096"
X Link 2025-12-19T14:03Z 437.6K followers, 64.1K engagements

"Software development will never be the same. I want you to watch this video: This is a spec-driven development environment. 100% of your time goes to writing specs and managing agents. 0% goes to writing the code"
X Link 2025-12-19T20:04Z 437.6K followers, 123.3K engagements

"This is a billion-dollar idea This is an open-source tool that detects signs of burnout in employees before it becomes a problem. The project is called On-Call Health and here's how it works: It combines two types of signals: [--]. Objective data from tools like PagerDuty Slack GitHub Linear and Jira. Think of incident volume after-hours pages number of tasks etc. [--]. Self-reported check-ins. These are short surveys asking engineers how they actually feel. Based on both the application computes a burnout risk score. For example if you have someone on your team who's been troubleshooting multiple"
X Link 2026-02-11T15:10Z 437.6K followers, 16.8K engagements

"Open source is the opposite of what every big AI lab is offering. I was using the Cline CLI for the first time and they offer the freedom you won't get with any of the proprietary agentic coding tools: It's open-source You aren't locked into a single model Bring your own API Key to use any model Over [--] million developers already use it Fastest-growing AI open-source project on GitHub Even better: Right now you can use two of the best open source models for free: MiniMax M2.5 and Kimi K2.5. To install it: npm install -g cline https://twitter.com/i/web/status/2022372717748765167"
X Link 2026-02-13T18:10Z 437.6K followers, 28.6K engagements

"Here is a completely free Computer Science curriculum for those with the discipline to go through it independently. The curriculum follows the same degree requirements as an undergraduate computer science major and consists of courses from Harvard MIT and Princeton among other colleges. The curriculum has four major sections: [--]. Introduction: Students can review this section and decide whether this fits them. [--]. Core: This section corresponds to the first three college years. [--]. Advanced: This section corresponds to the final year of a Computer Science curriculum in college. [--]. Final Project:"
X Link 2023-07-03T10:08Z 437.6K followers, 3.1M engagements

"This is the best laptop in the world. Period. M4 [--] GB RAM 1TB SSD [--] hours of battery life [--] inches [---] lbs. $1699 with the student discount. No competition"
X Link 2025-03-15T13:05Z 437.6K followers, 14.2M engagements

"Here is how you can give Claude Code access to any data that exists online. It's an easy way to make it 10x more powerful than it already is. For example: Use Claude Code to find open LinkedIn jobs in your area tailor your resume to them and apply for them automatically"
X Link 2026-01-29T16:10Z 437.6K followers, 178K engagements

"me every time claude code one-shots something i said it was impossible for AI to do"
X Link 2026-02-03T23:28Z 437.6K followers, 114.2K engagements

"A 100% automated QA team that works 24/7: [--]. Write your test in plain English [--]. AI generates the test cases [--]. Web agents execute the tests in parallel [--]. Live browser preview with everything that happens Try this for free in your project"
X Link 2026-02-08T15:04Z 437.6K followers, 53.1K engagements

"@warpdotdev released Oz and it might be the missing piece we all need: Oz is a cloud-based coding agent with built-in orchestration (the cloud part is important because you don't have to run it locally). You set up isolated environments point them at your repos and manage everything from a single panel. You can see everything running in a single place (instead of having to tab-switch between terminals): Local agents Cloud agents Scheduled agents Agents triggered from the API (Slack GitHub Actions etc.) You can control every agent from that dashboard. You can even fork cloud agents to run on"
X Link 2026-02-12T20:21Z 437.6K followers, [----] engagements

"big ai: software engineering is dead writing code is dead ai will write better binaries don't learn to write code everyone can build anything also big ai: "more than two colors is too hard""
X Link 2026-02-14T14:01Z 437.6K followers, 150.6K engagements

"your css intern is the only ai-proof role at your company big ai: software engineering is dead writing code is dead ai will write better binaries don't learn to write code everyone can build anything also big ai: "more than two colors is too hard" https://t.co/K7IVpKaDdt big ai: software engineering is dead writing code is dead ai will write better binaries don't learn to write code everyone can build anything also big ai: "more than two colors is too hard" https://t.co/K7IVpKaDdt"
X Link 2026-02-14T14:12Z 437.6K followers, [----] engagements

"Speculative decoding is not just "throughput optimization". In production systems the real impact of speculative decoding is on tail latency. Speculative decoding pairs your main model with a smaller faster draft model. The draft model proposes tokens; the main model verifies them in parallel. When the draft model predicts well you can skip serial generation steps and reduce latency. But there's a problem: A generic draft model wasn't trained on your data so they make many mistakes and the main model will reject most of their predictions. At this point any potential latency benefits are gone."
X Link 2026-02-14T15:35Z 437.6K followers, [----] engagements

"Ive been an iTerm2 user since forever but lately I keep hearing about Ghostty. Is it worth switching With Claude Code Im spending an obscene amount of time in the terminal so its worth considering any potential improvements"
X Link 2026-02-15T13:49Z 437.6K followers, 73.4K engagements

"Oh no They are stealing the stolen data OpenAI has warned US lawmakers that its Chinese rival DeepSeek is using unfair and increasingly sophisticated methods to extract results from leading US AI models to train the next generation of its breakthrough R1 chatbot https://t.co/siq46jRyM1 OpenAI has warned US lawmakers that its Chinese rival DeepSeek is using unfair and increasingly sophisticated methods to extract results from leading US AI models to train the next generation of its breakthrough R1 chatbot https://t.co/siq46jRyM1"
X Link 2026-02-15T02:23Z 437.6K followers, 41.3K engagements

""open"claw just became open as in "open"ai Peter Steinberger is joining OpenAI to drive the next generation of personal agents. He is a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people. We expect this will quickly become core to our Peter Steinberger is joining OpenAI to drive the next generation of personal agents. He is a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people. We expect this will quickly become core to our"
X Link 2026-02-15T23:42Z 437.6K followers, [----] engagements

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

@svpino
/creator/twitter::svpino