•twitter15877
listening to
timeline
15877
mentions
tracked
tracked
net sentiment
+24%
4998 positive
9734 neutral
1145 negative
volume & sentiment over time
Sep 30 – Oct 14, 2026 · Daily
sources
most discussed
—
active days
15
mentions
Showing 3926–3950 of 4998 mentions
-
• twitter 1w positiveHis weight-loss journey will genuinely inspire you 🫡🔥 https://t.co/aUeDVUukjc His weight-loss journey will genuinely inspire you 🫡🔥 https://t.co/aUeDVUukjcview source ↗
-
• twitter 1w positive> be me > watches f1 for the first time > it's actually a banger race apparently > friend calls me Kimi Antonelli of memory (learned who he is today, seems like a cool guy 🫡) > be me > watches f1 for the first time > it's actually a banger race apparently > friend calls me Kimi Antonelli of memory (learned who he is today, seems like a cool guy 🫡)view source ↗
-
• twitter 1w positiveLaunching https://t.co/Vgvb6xWGjl Read lafda's, write one or just yap, anything works. Still in beta. Feedback appreciated. https://t.co/w0vmpN8PcF Launching https://t.co/Vgvb6xWGjl Read lafda's, write one or just yap, anything works. Still in beta. Feedback appreciated. https://t.co/w0vmpN8PcFview source ↗
-
• twitter 1w positiveIf I had 1 month to build my Portfolio Project which will get me hired as an AI RAG Engineer, I would do these, starting from basics to advance: 1) Build your vectordb: Arrange a dataset with thousands of unstructured files. You can get it from any open source repository or platform. Now before chunking, you choose ty… If I had 1 month to build my Portfolio Project which will get me hired as an AI RAG Engineer, I would do these, starting from basics to advance: 1) Build your vectordb: Arrange a dataset with thousands of unstructured files. You can get it from any open source repository or platform. Now before chunking, you choose type of vectordb you require. Do you need a dedicated vectordb or pg vector extension? This should be a business impact based decision. If you need Vectordb + Crud operations in one database then go with pgvector extension. If you don't have crud operations in your business and require a high scalable vectordb then go with a dedicated vectordb. 2) Engineer the chunking strategy: The first and foremost strategy of any RAG pipeline is chunking. There are various types of chunking strategies like paragraph chunking, paragraph with overlap chunking, document layout aware based chunking, LLM based reasoning chunking, etc. For the files which contains text + images + tables, you can opt for either document layout aware based chunking or build a pipeline to process three different chunking for this three different requirements. 3) Build and Evaluate your first retriever like a real engineer: Choose your retriever strategy, either it could be semantic or keyword based or both. Now question is how you gonna choose? It could be usecase wise too, if that dataset has some keywords which is hard to retrieve semantically then use hybrid approch. In interview they will ask how do you know that semantically it will be harder to fetch those?? To answer this question like a True Real Engineer, you do experiments with this two different retriever strategies semantic vs hybrid. You take 1 eval dataset, and example queries, and give those examples queries to retrieval to fetch chunks from vectordb and then you evaluate those chunks with eval dataset with metrices like Precision, Recall, MRR, Hitrate, Context Recall and Context Relevance. Now you make decisions based on thview source ↗
-
• twitter 1w positiveDon't be affected by criticism. Positive or negative, take them in your stride. Know that the level of criticism rises with your worth. Don't be affected by criticism. Positive or negative, take them in your stride. Know that the level of criticism rises with your worth.view source ↗
-
• twitter 1w positiveWon first at the @supabase Select 2026 hackathon today with my project Fizz, a physical layer for agents! A huge thank you to @kiwicopple and @AntWilson and all of the amazing sponsors for making this a memorable event! Check out my project at: https://t.co/A6l71JzdOU (now in Waitlist!) Won first at the @supabase Select 2026 hackathon today with my project Fizz, a physical layer for agents! A huge thank you to @kiwicopple and @AntWilson and all of the amazing sponsors for making this a memorable event! Check out my project at: https://t.co/A6l71JzdOU (now in Waitlist!)view source ↗
-
• twitter 1w positiveOpus 5.5 on Max effort (Sound On): "A 24-second AI In Public motion reel in Remotion at 1920×1080, 60fps. Use our blue # 3D40FE, white, navy, and ice palette with the AI In Public logo and sparkle. Keep it fast, premium, and smooth with motion blur, glow, camera hits, HUD details, seamless transitions, kinetic type, t… Opus 5.5 on Max effort (Sound On): "A 24-second AI In Public motion reel in Remotion at 1920×1080, 60fps. Use our blue # 3D40FE, white, navy, and ice palette with the AI In Public logo and sparkle. Keep it fast, premium, and smooth with motion blur, glow, camera hits, HUD details, seamless transitions, kinetic type, terminal UI, neural visuals, particles, and a cinematic logo finale. Generate all suspenseful music and sound effects in code and sync every hit perfectly. End with “AI IN PUBLIC, learning AI, out loud.” Every frame and sound should be written in code." full prompt on here: https://t.co/Hp3qoiIww0 then read the full guide below: ↓view source ↗
-
• twitter 1w positiveCLAUDE CODE CAN NOW TURN HIGGSFIELD INTO A ONE-PERSON AI VIDEO STUDIO. install this 19 skills, connect Claude Code to Higgsfield, then tell Claude what you want to create. run /ugc-hot-girl and let it create the character, generate the visuals, and produce the Seedance video for you. Claude controls Higgsfield throu… CLAUDE CODE CAN NOW TURN HIGGSFIELD INTO A ONE-PERSON AI VIDEO STUDIO. install this 19 skills, connect Claude Code to Higgsfield, then tell Claude what you want to create. run /ugc-hot-girl and let it create the character, generate the visuals, and produce the Seedance video for you. Claude controls Higgsfield through Playwright. you just review the output and make changes when needed. this is useful if you want to make UGC ads for brands without spending hours doing everything manually. you could even turn the workflow into a service and sell the finished content. if you’re building toward the solo creator income path, save this. repo below 👇view source ↗
-
• twitter 1w positiveyou can have the best offer in your market and still lose the deal in the first 5 minutes a lead is hottest the second they hit submit reach them in 5 min and they still remember why they filled it out. reach them in 5 hours and they filled out 3 other forms and forgot yours the account isn't the problem. the re… you can have the best offer in your market and still lose the deal in the first 5 minutes a lead is hottest the second they hit submit reach them in 5 min and they still remember why they filled it out. reach them in 5 hours and they filled out 3 other forms and forgot yours the account isn't the problem. the response time is ->bookmarkview source ↗
-
• twitter 1w positiveFor someone who used to edit videos by just making just cuts and removing pauses, this feels like magic to me 🫡 - Opus 5.5 is wild - Full result: https://t.co/zThvLdeVw8 - This trailer edit is also edited by AI https://t.co/bTgcGFUJ0m For someone who used to edit videos by just making just cuts and removing pauses, this feels like magic to me 🫡 - Opus 5.5 is wild - Full result: https://t.co/zThvLdeVw8 - This trailer edit is also edited by AI https://t.co/bTgcGFUJ0mview source ↗
-
• twitter 1w positive1 month gym update 😋 : - Chest and triceps improved lil bit - feeling more stamina and immunity overall - less gastric issues I'm in hostel so need to increase the food and protein quantity Drop some tips chat https://t.co/9VKl6PxAar 1 month gym update 😋 : - Chest and triceps improved lil bit - feeling more stamina and immunity overall - less gastric issues I'm in hostel so need to increase the food and protein quantity Drop some tips chat https://t.co/9VKl6PxAarview source ↗
-
• twitter 1w positiveHermes Agent Community is Talented 👏 Hermes Agent Community is Talented 👏view source ↗
-
• twitter 1w positiveSend this Jev prompt to Opus 5.5 + any screenshot. It’ll: → Set up Jev → Turn your product screenshot into a 15-sec design motion video → Make future Opus 5.5 work faster + cheaper Try it first. Thank me later. https://t.co/Fcs0hDUg8M Send this Jev prompt to Opus 5.5 + any screenshot. It’ll: → Set up Jev → Turn your product screenshot into a 15-sec design motion video → Make future Opus 5.5 work faster + cheaper Try it first. Thank me later. https://t.co/Fcs0hDUg8Mview source ↗
-
• twitter 1w positiveYour AI agent stops working the second you close the tab. This one keeps going. Picture this: A permission slip needs filling out. You type one line into OpenMuse: "Complete the permission slip." Here's what happens next: → It turns that into a task with a visible plan → Opens the email attachment as a PDF → Asks yo… Your AI agent stops working the second you close the tab. This one keeps going. Picture this: A permission slip needs filling out. You type one line into OpenMuse: "Complete the permission slip." Here's what happens next: → It turns that into a task with a visible plan → Opens the email attachment as a PDF → Asks you for the details it needs → Fills in a copy of the form → Drafts a reply for you → Saves a receipt Then it waits. Nothing goes out until you review it. Need to jump in mid-task? Hit "Take control" and you're inside the same browser session the agent was using. That's OpenMuse from CopilotKit. Open source, MIT licensed, and running on iOS, Android, and web. 🚀 Save this video, you'll want it the next time your inbox piles up. Want the SOP? DM me.view source ↗
-
• twitter 1w positiveA moodboard shows what you liked. Video-to-Prompt shows you WHY it worked, then lets you swap the ingredients. Dreamina 2.0. #AIcreator A moodboard shows what you liked. Video-to-Prompt shows you WHY it worked, then lets you swap the ingredients. Dreamina 2.0. #AIcreatorview source ↗
-
• twitter 1w positiveHermes on your phone just got VERY real. 📱🔥 Bear’s been working on bringing the actual Hermes Desktop experience into the browser, and now he’s pushing it onto phones and tablets too. Your main machine keeps running Hermes. Your phone becomes the control surface. This could completely level up how you operate Hermes… Hermes on your phone just got VERY real. 📱🔥 Bear’s been working on bringing the actual Hermes Desktop experience into the browser, and now he’s pushing it onto phones and tablets too. Your main machine keeps running Hermes. Your phone becomes the control surface. This could completely level up how you operate Hermes away from your desk. 👇view source ↗
-
• twitter 1w positive“Any questions?” Silence. Great. Everyone understood. Or someone didn't understand and doesn't want to be the only one saying so. Or they disagree and have learnt that they are only reprimanded when they say so. Or the meeting has already run 20 minutes over! Try asking, “Which part have I not explained well?” You… “Any questions?” Silence. Great. Everyone understood. Or someone didn't understand and doesn't want to be the only one saying so. Or they disagree and have learnt that they are only reprimanded when they say so. Or the meeting has already run 20 minutes over! Try asking, “Which part have I not explained well?” You have made the same silence a little easier to break.view source ↗
-
• twitter 1w positiveJunior: My API request is showing up twice in DevTools. There’s only one call in my React code. If you know the reason, I already know you're a damn good developer. Junior: My API request is showing up twice in DevTools. There’s only one call in my React code. If you know the reason, I already know you're a damn good developer.view source ↗
-
• twitter 1w positiveHey @grok who would win if the United States, France & UK went to war against Russia, China & India without nuclear weapons? https://t.co/nqGbPwW5v4 Hey @grok who would win if the United States, France & UK went to war against Russia, China & India without nuclear weapons? https://t.co/nqGbPwW5v4view source ↗
-
• twitter 1w positiveLove this component https://t.co/yMU3YeaYPL https://t.co/4CWtghaB6J Love this component https://t.co/yMU3YeaYPL https://t.co/4CWtghaB6Jview source ↗
-
• twitter 1w positivemade this @vercel like animation https://t.co/8z3z5MgyeB made this @vercel like animation https://t.co/8z3z5MgyeBview source ↗
-
• twitter 1w positivewe added a profiler to Tern! literally playing easy mode, no excuses for slow software something something #RSI https://t.co/7D1Ma7rGIy we added a profiler to Tern! literally playing easy mode, no excuses for slow software something something #RSI https://t.co/7D1Ma7rGIyview source ↗
-
• twitter 1w positiveT3 Code (Nightly) is the best tool for building with agents rn: - BYO Claude, Codex, OpenCode, etc. - Codex tier desktop app quality & design - Easy cloud agent setup with T3 Connect - Support to bring multiple subs - Native delegation between providers + so many other things T3 Code (Nightly) is the best tool for building with agents rn: - BYO Claude, Codex, OpenCode, etc. - Codex tier desktop app quality & design - Easy cloud agent setup with T3 Connect - Support to bring multiple subs - Native delegation between providers + so many other thingsview source ↗
-
• twitter 1w positiveVery underrated tip from @karpathy. Just tried ASD-STE100 rule for model responses, and I found it to be surprisingly good at making explanations clearer. It can be bit too strict, so I agree that "80% of the way to ASD-STE100" feels like a better default. A prompt you can try with your agent: "Use ASD-STE100 as a g… Very underrated tip from @karpathy. Just tried ASD-STE100 rule for model responses, and I found it to be surprisingly good at making explanations clearer. It can be bit too strict, so I agree that "80% of the way to ASD-STE100" feels like a better default. A prompt you can try with your agent: "Use ASD-STE100 as a guide for your responses. Write short sentences, use active voice, and keep terminology consistent. Relax the vocabulary rules when they make explanations awkward. Preserve technical precision and uncertainty." @0xpili_ also made a nice skill for it: https://t.co/UODgEeQy6zview source ↗
-
• twitter 1w positive@poteto I'm wondering how you manage your "kitchen" when it comes to product decisions. My biggest bottleneck is myself needing to be involved with product direction. Of course I want the app faster and more secure and more accessible and fixing user-submitted big reports and I can fire off agents to handle all that a… @poteto I'm wondering how you manage your "kitchen" when it comes to product decisions. My biggest bottleneck is myself needing to be involved with product direction. Of course I want the app faster and more secure and more accessible and fixing user-submitted big reports and I can fire off agents to handle all that autonomously, but the things that really move the needle is whether how I think about the evolution of my product and its features. How do you think about that (and how do you coordinate with the other people on your team)? I've watched you talk about you process a lot in interviews and things and it maps well to things I've been doing also (and I borrowed the CLI idea which has been fantastic). But nobody seems to have asked you about how you think about and manage the product side of things so I'd love to hear more about that! Thanks!view source ↗