•twitter4892
listening to
timeline
4892
mentions
tracked
tracked
net sentiment
+24%
1527 positive
2999 neutral
366 negative
volume & sentiment over time
Sep 30 – Oct 5, 2026 · Daily
sources
most discussed
—
active days
6
mentions
Showing 551–575 of 1527 mentions
-
• twitter 1d positiveAnother look at Blyth AI Mobile. I’m really happy with the overall design so far. There are still a few details I’m refining before release, but we’re getting close. What would you change or improve? Any feedback is welcome. https://t.co/L1Ybpd7rEc Another look at Blyth AI Mobile. I’m really happy with the overall design so far. There are still a few details I’m refining before release, but we’re getting close. What would you change or improve? Any feedback is welcome. https://t.co/L1Ybpd7rEcview source ↗
-
• twitter 1d positiveit's easy to fall into the trap of micromanaging your agents instead of correcting the environment that shapes its behavior. pstack 0.15.9 now ships a new /correct skill: if you keep correcting agents for the same mistakes, it finds the pattern and fixes it with architecture, types, and checks. i like to imagine tha… it's easy to fall into the trap of micromanaging your agents instead of correcting the environment that shapes its behavior. pstack 0.15.9 now ships a new /correct skill: if you keep correcting agents for the same mistakes, it finds the pattern and fixes it with architecture, types, and checks. i like to imagine that commits rapidly flowing into the codebase are like a bonsai tree rapidly growing in whatever direction it pleases. what you want to do is not to give in to the chaos, but tame it intentionally with the right constraints 0.15.9 also has improvements to: • /architect: now includes instructions on designing agent friendly architecture • a new /benchmark-checklist skill based on @brendangregg's benchmarking checklist here's a prompt you can give to grok bot to start architecting a better codebase with pstack: /poteto-mode create a new Project agent to refactor and rearchitect my repo so that our architecture is more agent friendly. it should use both the /correct and /architect skill to look through past commits and find the most common pitfalls that agents fall into, whether its with code quality, performance, or bugs. if relevant you can also use the /recall skill to find past conversations for context. it should make a high level plan for me to review. it should also answer its own open questions by prototyping rather than defer to me. come back to me with a plan backed by real data. once i approve the plan, use either the stack or full autopilot playbook (ask me which i want) to executeview source ↗
-
• twitter 1d positivewe're in the third phase of AI for business phase 1: chat chatgpt and claude. you ask, it answers phase 2: connected chat connectors, integrations, a bit of memory. it can see your stuff phase 3: agents with their own computer openclaw was first. then grok bot, muse, and now openai's dots tool calls, computer use, … we're in the third phase of AI for business phase 1: chat chatgpt and claude. you ask, it answers phase 2: connected chat connectors, integrations, a bit of memory. it can see your stuff phase 3: agents with their own computer openclaw was first. then grok bot, muse, and now openai's dots tool calls, computer use, learning your business over time, and taking actions for you this is openclaw going mainstream phase 4 is next: the multiplayer AI workspace your team and your agents in one platform, working on the same stuff together openai's space is the first look at it. people, chatgpt and dots building on shared knowledge the gap is the agents. one dot per person for now. openai showed specialist dots for roles like invoice processing, procurement and support, but they're enterprise pilots I've been building this by hand for my client builds: - the agents live in the team's slack, where the work already happens - every step that leaves the building gets a human approval - one always-on agent for the daily work, and a coding agent for building the software the labs are about to make that the default the businesses that already know which roles to hand off will move fastestview source ↗
-
• twitter 1d positive5 AM Good morning developers! 👨💻☕ 5 AM Good morning developers! 👨💻☕view source ↗
-
• twitter 1d positiveExciting to see this launch! ⭐️ What decision-makings are you already implementing in real projects? 🎉 Exciting to see this launch! ⭐️ What decision-makings are you already implementing in real projects? 🎉view source ↗
-
• twitter 1d positiveWhat are you building? 🧠 AI 🚀 SaaS 📱 Mobile 🌐 Web 🛠️ Dev tools 🎨 Design Reply with your category and one line 👇 https://t.co/ofJIeEOhwB What are you building? 🧠 AI 🚀 SaaS 📱 Mobile 🌐 Web 🛠️ Dev tools 🎨 Design Reply with your category and one line 👇 https://t.co/ofJIeEOhwBview source ↗
-
• twitter 1d positiveI don’t think people realize how much their location can affect their life. The people. The morals. The weather. The Cost. If you’ve never traveled, I’d highly recommend. I don’t think people realize how much their location can affect their life. The people. The morals. The weather. The Cost. If you’ve never traveled, I’d highly recommend.view source ↗
-
• twitter 1d positivethe Actors model (aka Durable Execution) has been quietly taking over modern development practices for years so many benefits compared to conventional web services Durable Objects / Temporal / Step Functions / Trigger / Restate and many more but agents make it the standard, required pattern - build your agents any o… the Actors model (aka Durable Execution) has been quietly taking over modern development practices for years so many benefits compared to conventional web services Durable Objects / Temporal / Step Functions / Trigger / Restate and many more but agents make it the standard, required pattern - build your agents any other way and they simply don't work and / or cost a fortune (because agent sessions can run for many days) it used to be a niche "durable workflow" scenario that was only truly needed in a small fraction of already complex applications but now it's the main primitive powering "the new apps" literally every agent thread (including user facing chats and background agent executions) is a durable workflow this calls for further simplification most users aren't going to wrap their heads around Temporal / alternatives. it's a truly beautiful set of ideas, but for better or worse it's too abstract to go mainstream, just like FP so - simplify we should that's why we're building Serverless Agents https://t.co/kxBKGd6Km4view source ↗
-
• twitter 1d positive🚨 𝟲𝗛𝗥𝗦 & 𝟯 𝗦𝗣𝗢𝗧𝗦 𝗟𝗘𝗙𝗧! 𝗖𝗟𝗔𝗜𝗠 𝗬𝗢𝗨𝗥 𝗙𝗥𝗘𝗘 𝗙𝗨𝗟𝗟 𝗙𝗨𝗡𝗡𝗘𝗟 𝗕𝗨𝗜𝗟𝗗 𝗔𝗡𝗗 𝗚𝗘𝗧 𝗦𝗘𝗘𝗡, 𝗚𝗘𝗧 𝗟𝗘𝗔𝗗𝗦, 𝗚𝗘𝗧 𝗣𝗔𝗜𝗗. 𝗚𝗢𝗡𝗘 𝗧𝗢𝗡𝗜𝗚𝗛𝗧. 🚨 → https://t.co/Ppmk5h9rw3 🚨 𝟮 𝗙𝗢𝗨𝗡𝗗𝗘𝗥𝗦 𝗔𝗟𝗥𝗘𝗔𝗗𝗬 𝗜𝗡. 𝗢𝗡𝗘 𝗣𝗥𝗢𝗗𝗨𝗖𝗧 𝗣𝗛𝗢𝗧𝗢 𝗕𝗘𝗖𝗢𝗠𝗘𝗦 𝗨𝗚𝗖 𝗩𝗜𝗗𝗘𝗢𝗦, 𝗖𝗔𝗥𝗢𝗨𝗦𝗘𝗟𝗦, 𝗔𝗡𝗗 𝗦𝗧𝗢𝗥𝗜𝗘𝗦 𝗧𝗛𝗔𝗧 𝗣𝗢𝗦𝗧 𝗧𝗛𝗘 𝗠𝗢𝗠𝗘𝗡𝗧 𝗬𝗢𝗨 𝗔𝗣𝗣𝗥𝗢𝗩𝗘. 𝗡𝗢 𝗖𝗔𝗠𝗘𝗥𝗔. 𝗡𝗢 𝗘𝗗𝗜𝗧𝗢𝗥. 𝗬𝗢𝗨𝗥 𝗙𝗘𝗘𝗗 𝗦𝗧𝗔𝗬𝗦 𝗔𝗟𝗜𝗩𝗘 𝗪𝗛𝗜𝗟𝗘 𝗬𝗢𝗨'𝗥𝗘 𝗕𝗨𝗦𝗬. 🚨→… 🚨 𝟲𝗛𝗥𝗦 & 𝟯 𝗦𝗣𝗢𝗧𝗦 𝗟𝗘𝗙𝗧! 𝗖𝗟𝗔𝗜𝗠 𝗬𝗢𝗨𝗥 𝗙𝗥𝗘𝗘 𝗙𝗨𝗟𝗟 𝗙𝗨𝗡𝗡𝗘𝗟 𝗕𝗨𝗜𝗟𝗗 𝗔𝗡𝗗 𝗚𝗘𝗧 𝗦𝗘𝗘𝗡, 𝗚𝗘𝗧 𝗟𝗘𝗔𝗗𝗦, 𝗚𝗘𝗧 𝗣𝗔𝗜𝗗. 𝗚𝗢𝗡𝗘 𝗧𝗢𝗡𝗜𝗚𝗛𝗧. 🚨 → https://t.co/Ppmk5h9rw3 🚨 𝟮 𝗙𝗢𝗨𝗡𝗗𝗘𝗥𝗦 𝗔𝗟𝗥𝗘𝗔𝗗𝗬 𝗜𝗡. 𝗢𝗡𝗘 𝗣𝗥𝗢𝗗𝗨𝗖𝗧 𝗣𝗛𝗢𝗧𝗢 𝗕𝗘𝗖𝗢𝗠𝗘𝗦 𝗨𝗚𝗖 𝗩𝗜𝗗𝗘𝗢𝗦, 𝗖𝗔𝗥𝗢𝗨𝗦𝗘𝗟𝗦, 𝗔𝗡𝗗 𝗦𝗧𝗢𝗥𝗜𝗘𝗦 𝗧𝗛𝗔𝗧 𝗣𝗢𝗦𝗧 𝗧𝗛𝗘 𝗠𝗢𝗠𝗘𝗡𝗧 𝗬𝗢𝗨 𝗔𝗣𝗣𝗥𝗢𝗩𝗘. 𝗡𝗢 𝗖𝗔𝗠𝗘𝗥𝗔. 𝗡𝗢 𝗘𝗗𝗜𝗧𝗢𝗥. 𝗬𝗢𝗨𝗥 𝗙𝗘𝗘𝗗 𝗦𝗧𝗔𝗬𝗦 𝗔𝗟𝗜𝗩𝗘 𝗪𝗛𝗜𝗟𝗘 𝗬𝗢𝗨'𝗥𝗘 𝗕𝗨𝗦𝗬. 🚨→ https://t.co/Ppmk5h9rw3 🚨 𝟯 𝗦𝗣𝗢𝗧𝗦. 𝟲 𝗛𝗢𝗨𝗥𝗦. 𝗖𝗟𝗔𝗜𝗠𝗘𝗗 𝗢𝗥 𝗠𝗜𝗗𝗡𝗜𝗚𝗛𝗧 𝗣𝗔𝗖𝗜𝗙𝗜𝗖, 𝗪𝗛𝗜𝗖𝗛𝗘𝗩𝗘𝗥 𝗖𝗢𝗠𝗘𝗦 𝗙𝗜𝗥𝗦𝗧. 🚨 → https://t.co/Ppmk5h9rw3 2 founders in. 3 spots left. 6 hours. Let's start where every funnel starts. Getting seen. Most people's content dies the same way. They post hard for a week, life gets busy, the feed goes quiet, and the right people stop finding them. It isn't that they ran out of ideas. They ran out of time to film, edit, and post. Step 1 of the Full Funnel Build kills that problem. No camera. No editor. You take one product photo and it becomes UGC videos, carousels, and stories. Blotato posts them the moment you approve. So your feed stays alive on the days you're too busy to think about it. And you don't build it with code. You describe it in plain English in the Agent OS chat. The agents build it. You test it with a bit of back and forth. I used to spend one or two hours wiring a single automation in n8n just to film a tutorial. Now it's one sentence and a bit of testing. That's step 1. Steps 2 and 3 turn those views into inquiries and those inquiries into paid invoices. All three, free, for the first 5 who join today. 3 spots left. 6 hours. 👉 Claim one of the last 3 Full Funnel Builds → https://t.co/Ppmk5h9rw3 🎯 𝗧𝗛𝗘 𝗙𝗨𝗟𝗟 𝗙𝗨𝗡𝗡𝗘𝗟 𝗕𝗨𝗜𝗟𝗗 (𝗙𝗥𝗘𝗘 𝗙𝗢𝗥 𝗧𝗛𝗘 𝗙𝗜𝗥𝗦𝗧 𝟱): ✅ Step 1: Get seen ... one product photo becomes UGC videos, carousels, and stories, posted by Blotato the moment you approve ✅ Step 2: Get the lead ... one Claude prompt builds the site all your content points to ✅ Step 3: Get paid ... AI follow-up answers new leads, chases quiet quotes, nudges late invoices, rebooks no-shows 💎 𝗬𝗢𝗨𝗥 𝗦𝗘𝗔𝗧 𝗖𝗢𝗩𝗘𝗥𝗦 𝗧𝗛𝗘 𝗥𝗘𝗦𝗧: ✅view source ↗
-
• twitter 1d positiveHurt me to not include: - Untrue by Burial - Hopelessness by Anohni - Mosaik by Siriusmo - Los Thuthanaka s/t - DJ E by Chuquimamani-Condori - BUBBA by Kaytranada - This Is Happening by LCD Soundsystem Honorable mentions: - Hypochondriac by Brakence - Oi Oi Oi by Boys Noize - James Blake s/t - R Plus Seven by Oneohtri… Hurt me to not include: - Untrue by Burial - Hopelessness by Anohni - Mosaik by Siriusmo - Los Thuthanaka s/t - DJ E by Chuquimamani-Condori - BUBBA by Kaytranada - This Is Happening by LCD Soundsystem Honorable mentions: - Hypochondriac by Brakence - Oi Oi Oi by Boys Noize - James Blake s/t - R Plus Seven by Oneohtrix Point Never - Government Plates by Death Grips - TNGHT s/t - Rap Album Two by Jonwayne - Couch Potato by Bobby Raps x Corbin - 09/17 2007 by Danger - RAT by Deadmau5 - Jai Paul's album leak thing - Aa by Baauer - Maybe You've Been Brainwashed Too by New Radicals - Causers of This by Toro y Moi - Atrocity Exhibition by Danny Brown - The Eulogy by Cakes da Killa - Rest by Charlotte Gainsbourg So much shit I'm still forgetting... Music is so cool.view source ↗
-
• twitter 1d positiveParis is burning again. Students “protesting school conditions” are setting fires, attacking cops, and beating an officer half to death on camera. Macron’s France spent decades on unfettered immigration and open borders. The bill is overcrowded schools, no-go suburbs, and streets that look like a war zone. This is the … Paris is burning again. Students “protesting school conditions” are setting fires, attacking cops, and beating an officer half to death on camera. Macron’s France spent decades on unfettered immigration and open borders. The bill is overcrowded schools, no-go suburbs, and streets that look like a war zone. This is the future the open-borders crowd wants for America. Stop it now. https://t.co/eDudmwnZ0tview source ↗
-
• twitter 1d positiveART ART ART !!!! 🔥🔥🔥 https://t.co/nm6t2jVPr4 ART ART ART !!!! 🔥🔥🔥 https://t.co/nm6t2jVPr4view source ↗
-
• twitter 1d positiveQ4 meeting done with my https://t.co/ecKp6beLxo team. Our company is fast growing. Here are some great additions to our team: -2 new video editors -2 website and AI experts -1 media buyer -1 content writer Let’s go! https://t.co/SmTA0po7GY Q4 meeting done with my https://t.co/ecKp6beLxo team. Our company is fast growing. Here are some great additions to our team: -2 new video editors -2 website and AI experts -1 media buyer -1 content writer Let’s go! https://t.co/SmTA0po7GYview source ↗
-
• twitter 1d positiveCurrent agents I'm using/testing: @interaction Poke <3 @NousResearch Hermes <3 @wajo_ai Fo <3 @Instinct fine and free @pallyai promising @CueAgents struggles w workarounds @wabi Mwah amazing actually improves your life without asking Willing to test more Current agents I'm using/testing: @interaction Poke <3 @NousResearch Hermes <3 @wajo_ai Fo <3 @Instinct fine and free @pallyai promising @CueAgents struggles w workarounds @wabi Mwah amazing actually improves your life without asking Willing to test moreview source ↗
-
• twitter 1d positiveFor whatever it's worth, there is no other platform in the world where the CEO of the third largest company in the world personally responds to a request from his biggest competitor about listing on their platform. Competition aside, let's take a moment to appreciate what a special platform X is. I keep telling my fr… For whatever it's worth, there is no other platform in the world where the CEO of the third largest company in the world personally responds to a request from his biggest competitor about listing on their platform. Competition aside, let's take a moment to appreciate what a special platform X is. I keep telling my friends this. X has the highest % of influential people actually running their own accounts. Trust me, this is actually Sundar typing on his Pixel, replying to Peter. What an amazing place.view source ↗
-
• twitter 1d positivePractice makes perfect Do so much …. that they talk about you so much …. that you eventually can’t hear them …. Understand? https://t.co/mddTl3DUD8 Practice makes perfect Do so much …. that they talk about you so much …. that you eventually can’t hear them …. Understand? https://t.co/mddTl3DUD8view source ↗
-
• twitter 1d positiveKeep believing that your hard work is leading you towards an incredible future. Keep believing that your hard work is leading you towards an incredible future.view source ↗
-
• twitter 1d positiveSmallest started with the same philosophy. Our name says it all. We’ve always believed that smaller can be better, faster, more efficient, and more accessible. Everything we build is an attempt to push that thesis further. Smallest started with the same philosophy. Our name says it all. We’ve always believed that smaller can be better, faster, more efficient, and more accessible. Everything we build is an attempt to push that thesis further.view source ↗
-
• twitter 1d positiveit is 3rd of October, so day 3 of 31 days of October! and here’s another amazing thing @october_dev can do for you no matter where you are, you can orchestrate your agents! I’m at the airport and I can run a fleet of agents on the cloud or on my laptop back home! Go try October if you haven’t! 🎃🎃 it is 3rd of October, so day 3 of 31 days of October! and here’s another amazing thing @october_dev can do for you no matter where you are, you can orchestrate your agents! I’m at the airport and I can run a fleet of agents on the cloud or on my laptop back home! Go try October if you haven’t! 🎃🎃view source ↗
-
• twitter 1d positiveI fixed the Codex app, thoughts? https://t.co/c89XMoFZTz I fixed the Codex app, thoughts? https://t.co/c89XMoFZTzview source ↗
-
• twitter 1d positiveAnd now that you don't have to be perfect, you can be good. East of Eden is now playing. https://t.co/8ouUSn5kId And now that you don't have to be perfect, you can be good. East of Eden is now playing. https://t.co/8ouUSn5kIdview source ↗
-
• twitter 1d positivenow that i’m in consumer, i find myself constantly asking how to make a product as good as tap tap revenge now that i’m in consumer, i find myself constantly asking how to make a product as good as tap tap revengeview source ↗
-
• twitter 1d positiveI created a ChatGPT hook which can convert computer use actions into a video with cursor movements Its super useful for testing ChatGPT can test your product and attach this video with your PRs for you to review later You can even record demo videos for your customers or documentation It has no overhead at all as … I created a ChatGPT hook which can convert computer use actions into a video with cursor movements Its super useful for testing ChatGPT can test your product and attach this video with your PRs for you to review later You can even record demo videos for your customers or documentation It has no overhead at all as it uses the existing screenshots that the app already takes and combines it with the raw ghost cursor coordinates finally rendering it into a video as soon as the actions end let me know if you want something like this to be a part of your workflowview source ↗
-
• twitter 1d positiveStop Chatting With Claude. Start Working With It Like a Senior Engineer. Most people open Claude, type a one-line question, copy the answer, and move on. That works for small things. It is also why they think AI coding is overhyped. A senior engineer doesn't work from one-line questions. They need context, a clear go… Stop Chatting With Claude. Start Working With It Like a Senior Engineer. Most people open Claude, type a one-line question, copy the answer, and move on. That works for small things. It is also why they think AI coding is overhyped. A senior engineer doesn't work from one-line questions. They need context, a clear goal, and a way to know when the job is done. Claude is the same. These seven habits change the results. 1. Onboard it like a new hire Doc-Folder Create a CLAUDE.md file in your project root. Claude Code reads it at the start of every session. Your stack and folder structure Commands to build, test, and lint Conventions that differ from the defaults Gotchas that only your team knows Keep it short. A long file buries the rules that matter. 2. Ask for a plan before any code Never start with "build this feature." First, ask Claude to read the relevant code Then ask for a plan, with no code yet Review the plan and correct it Only then let it implement Fixing a plan takes two minutes. Fixing the wrong implementation takes two hours 3. Give it a way to check its own work This is the biggest upgrade you can make. Write the tests first, or ask Claude to write them Tell it to run tests, type checks, and the linter after every change For UI work, give it a screenshot or mockup to compare against Without a target, Claude gives you code that only looks right. With one, it keeps going until the code works. 4. Point at real files, not descriptions Reference the exact file instead of explaining where things live Show it an existing pattern and say "follow this" Paste the real error log, not your summary of it 5. Keep sessions clean Long, messy conversations produce worse output. One task per session Clear the context when you switch tasks If a session goes off track, start fresh instead of arguing with it 6. Work like a tech lead Your job changes from typing code to directing work. Break the work into clear, small tasks Run separate sessions for sepview source ↗
-
• twitter 1d positiveagents are the new kind of workload a 3rd type the 1st is backend APIs - they serve HTTP requests and go to managed containers or serverless functions the 2nd is frontends - they go to CDNs + edge functions agent sessions are the new, 3rd type. they need to run durably for hours or days the closest approximation… agents are the new kind of workload a 3rd type the 1st is backend APIs - they serve HTTP requests and go to managed containers or serverless functions the 2nd is frontends - they go to CDNs + edge functions agent sessions are the new, 3rd type. they need to run durably for hours or days the closest approximation before agents were perhaps Step Functions on AWS and alternatives that made use of durable execution. but they never became a mainstream "standard form" like a web service or a frontend bundle now with agents it's inevitably going to happen you describe it in code, just like an API or a web app you deploy it, with versions rollbacks etc you start sessions which can run indefinitely this can be built on Step Functions or Temporal or Trigger or Restate or any other durable execution engine of course but most users will not go that deep. it needs to be simplified - it wants to - it all but screams "simplify me!" so we built Serverless Agents define your agent Typescript. intuitive SDK. it feels just like building web apps or APIs, just one "app" to think of. blazing-fast deploy. try it - this is the future! https://t.co/kxBKGd6Km4view source ↗