•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 2476–2500 of 4998 mentions
-
• twitter 1w positiveBending Spoons doesn't let team leads pick their own hires. Luca Ferrari says hiring managers have the wrong incentives: 1. They want the seat filled fast 2. They take the first adequate person 3. They prefer experienced people who need no coaching A central team hires for everyone. It sees 800,000 applications a yea… Bending Spoons doesn't let team leads pick their own hires. Luca Ferrari says hiring managers have the wrong incentives: 1. They want the seat filled fast 2. They take the first adequate person 3. They prefer experienced people who need no coaching A central team hires for everyone. It sees 800,000 applications a year. Full breakdown here: https://t.co/rODle2JEOyview source ↗
-
• twitter 1w positiveThe kind of design that gets the users excited and gets the owners downloads.✨ One of the recent app designs we have worked on. https://t.co/D98J1hvnUt The kind of design that gets the users excited and gets the owners downloads.✨ One of the recent app designs we have worked on. https://t.co/D98J1hvnUtview source ↗
-
• twitter 1w positiveHermes is getting a whole lot easier to manage. Hermes merged 142 PRs on October 5. 3 changes worth knowing about: • Desktop now warns you before a subscription hits its usage wall and shows when rate-limited providers reset • Plugin skills now show up as /plugin:skill commands across Desktop, TUI, and CLI, and you… Hermes is getting a whole lot easier to manage. Hermes merged 142 PRs on October 5. 3 changes worth knowing about: • Desktop now warns you before a subscription hits its usage wall and shows when rate-limited providers reset • Plugin skills now show up as /plugin:skill commands across Desktop, TUI, and CLI, and you can stack multiple skills together • Desktop now has one dedicated Plugins settings area where every plugin can have its own settings page Luke has the full 142 below 👇view source ↗
-
• twitter 1w positiveStop overpaying for api tokens, do this instead 🚨 so i switched to deepseek v4 flash…free on codex/claude paired with glm 5.3 flash as subagents with antseed free. no card. no account. 200+ providers on the network. [it takes 3 mins to set up — here is how] 1. download antseed (https://t.co/m8Cp4Kdr3U) 2. connect c… Stop overpaying for api tokens, do this instead 🚨 so i switched to deepseek v4 flash…free on codex/claude paired with glm 5.3 flash as subagents with antseed free. no card. no account. 200+ providers on the network. [it takes 3 mins to set up — here is how] 1. download antseed (https://t.co/m8Cp4Kdr3U) 2. connect codex/claude code/hermes etc 3. pick a provider, that’s it It is anonymous, happy tokenmaxxingview source ↗
-
• twitter 1w positiveI've been reunited with the XPS14! Thanks again to the kind people at @Lufthansa_USA for helping. I'll take better care next time I'm jetlagged out of my mind 🫣🙏 https://t.co/5dWP4oh3jm I've been reunited with the XPS14! Thanks again to the kind people at @Lufthansa_USA for helping. I'll take better care next time I'm jetlagged out of my mind 🫣🙏 https://t.co/5dWP4oh3jmview source ↗
-
• twitter 1w positiveHow to win Google AI: How to win Google AI:view source ↗
-
• twitter 1w positivethis is f**king insane every AI model in one Telegram chat and the balance still says $0.00 before this, every question from my phone meant going back to the laptop and opening the desktop app. Antseed runs on the computer and brings the same agent into a private Telegram bot. Telegram is just the remote. [the whol… this is f**king insane every AI model in one Telegram chat and the balance still says $0.00 before this, every question from my phone meant going back to the laptop and opening the desktop app. Antseed runs on the computer and brings the same agent into a private Telegram bot. Telegram is just the remote. [the whole setup is in the 70 seconds above, here is the short version] 1. open Antseed → Apps → Connect on Telegram Bot 2. make a bot with BotFather and paste the token. it stays encrypted on your machine 3. send /model and pick: Claude Opus 5, GPT 5.6, Kimi K3, DeepSeek V4 or a free-tier model I kept GLM 4.5 Flash. free tier, 0 sign-up, 0 email. any model in the network. one chat. pay per use or pay nothing. download is in the post above, desktop only. Antseed funded my review so I could show what I found useful. first comment or in the body, up to the client Download Antseed - Desktop only https://t.co/iGBBNApcDb · @antseed · https://t.co/dB5ps1Z8qVview source ↗
-
• twitter 1w positiveIn August, https://t.co/gRgoAxK9Wc took over X. I shipped a daily clone on Aug 23 - founders bid free and paid, it made me my first dollars. Then the trend died. A 24-hour board has no long game. So I rebuilt it as https://t.co/vRgiG7N6Fo: - weekly launches, - permanent pages that rank on Google, - dofollow back… In August, https://t.co/gRgoAxK9Wc took over X. I shipped a daily clone on Aug 23 - founders bid free and paid, it made me my first dollars. Then the trend died. A 24-hour board has no long game. So I rebuilt it as https://t.co/vRgiG7N6Fo: - weekly launches, - permanent pages that rank on Google, - dofollow backlinks, - automated audits with fix cards for your coding agent. Pro adds weekly scans, competitor monitoring, and your AI-found #1 fix. 15 Pro seats at $9/m for solo founders. Ship it on → https://t.co/vRgiG7N6Fo #launchday #outbid #buildinpublic #shippedit #nhrdevview source ↗
-
• twitter 1w positiveA good person to meet A good person to meetview source ↗
-
• twitter 1w positiveClaude is getting too good at minimal UI animations https://t.co/TlKs4jClMR Claude is getting too good at minimal UI animations https://t.co/TlKs4jClMRview source ↗
-
• twitter 1w positiveObviously simple tools like this aren't killing Clay but this thing is sick for a few reasons It lets you scope tools to different agent sets This effectively gives you a composable code mode MCP You can have a dedicated track set up for your SEO, go-to-market, reply agent, you name it Additionally it just work… Obviously simple tools like this aren't killing Clay but this thing is sick for a few reasons It lets you scope tools to different agent sets This effectively gives you a composable code mode MCP You can have a dedicated track set up for your SEO, go-to-market, reply agent, you name it Additionally it just works. The uptime and speed is unbelievable and it's because of the star-studded team behind it Plus they just do a great job of taking tool requests from users. My stack now looks like this as a result Quick enrich -> AI ark -> Treg for all last-mile enrichment -> And if I'm in a real pinch then custom scraping agents powered by Serper, tinyfish, and exa.view source ↗
-
• twitter 1w positiveCareer plot twist: I got laid off 💀 After a year at Bicycle AI, my role including others were impacted by a restructuring. Not the plan, but i’m using the pause to figure out what work actually excites me while concentrating on my fitness goals & competitions (more on IG) ✨ Career plot twist: I got laid off 💀 After a year at Bicycle AI, my role including others were impacted by a restructuring. Not the plan, but i’m using the pause to figure out what work actually excites me while concentrating on my fitness goals & competitions (more on IG) ✨view source ↗
-
• twitter 1w positiveMeet Mistral Large 4, aka Le Chonk. • 1T parameters, natively multimodal. 49B active. It is the best open weights model from US or Europe on aggregated benchmarks. • State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding. … Meet Mistral Large 4, aka Le Chonk. • 1T parameters, natively multimodal. 49B active. It is the best open weights model from US or Europe on aggregated benchmarks. • State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding. • Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure. • Available to all via API today. Working with cybersecurity partners privately. Open weights release end of October.view source ↗
-
• twitter 1w positiveThe kind of impact that makes all the effort worth it :) I have always tried to keep things practical and focus on fundamentals, because they outlive frameworks, tools, and trends. Enrollments are now open for Applied AI - Cohort 6, in case you are interested. If you are already building AI products, or want to devel… The kind of impact that makes all the effort worth it :) I have always tried to keep things practical and focus on fundamentals, because they outlive frameworks, tools, and trends. Enrollments are now open for Applied AI - Cohort 6, in case you are interested. If you are already building AI products, or want to develop the intuition and mental models needed to build them well and run them in production, I think you will find this cohort valuable. The goal is not to teach you a bunch of tools, but to help you build mental models that stay useful even as the tools keep changing. Cohort 6th starts on 14 November; details, curriculum, testimonials - https://t.co/JhtIHMrLbgview source ↗
-
• twitter 1w positivethis is pure f*cking treasure A Stanford AI research group found how to orchestrate Claude Opus 5.5 and GPT-6.1 Sol together, and it completely breaks the search scaffolding wall most developers try to combine frontier models by chaining prompts in one context window or hardcoding rigid evolutionary loops. you burn t… this is pure f*cking treasure A Stanford AI research group found how to orchestrate Claude Opus 5.5 and GPT-6.1 Sol together, and it completely breaks the search scaffolding wall most developers try to combine frontier models by chaining prompts in one context window or hardcoding rigid evolutionary loops. you burn tokens on context bloat and freeze models into fixed search rules that out-date themselves within 5 iterations Stanford's architecture eliminates human-designed search scaffolding and splits the cognitive stack: > search director: Claude Opus 5.5 plans the search, querying a persistent idea-graph where MAP-Elites and MCTS reduce to single Cypher queries > candidate proposer: GPT-6.1 Sol explores high-entropy code variations inside isolated execution sandboxes > zero context bleeding: sessions reset after each iteration to prevent prompt bloat while the graph stores lineage > evaluator isolation: candidate code never touches the scoring process, stopping reward hacking and test leakage cold > meta-agent distillation: an offline meta-pass extracts verified lessons and updates search guidance across branches the benchmark metrics outclass standard discovery baselines: > 3.2x lower model spend than fixed evolutionary frameworks > 1st place rank across 7 competitive AtCoder heuristic contests against human competitors > beats published SOTA on Anthropic kernel builder and 11 mathematical optimization tasks > 0% reward hacks: candidate-controlled metric tampering eliminated Opus directs the search graph. Sol explores the code you stop writing rigid search harnesses. you let frontier models own the discovery loopview source ↗
-
• twitter 1w positiveEvery time you let Opus 5.5 work on a project overnight, have it vibe-code a quick 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗺𝗮𝗽 like this first. In the morning, you'll see at a glance what's done and what's left. Then give it 𝗮 𝘀𝘂𝗯𝗮𝗴𝗲𝗻𝘁 𝘁𝗵𝗮𝘁 𝗼𝗻𝗹𝘆 𝗱𝗿𝗮𝘄𝘀 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗺𝗮𝗽𝘀: → Name it project-map, set its effort to medium, and preload a design skill → The fir… Every time you let Opus 5.5 work on a project overnight, have it vibe-code a quick 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗺𝗮𝗽 like this first. In the morning, you'll see at a glance what's done and what's left. Then give it 𝗮 𝘀𝘂𝗯𝗮𝗴𝗲𝗻𝘁 𝘁𝗵𝗮𝘁 𝗼𝗻𝗹𝘆 𝗱𝗿𝗮𝘄𝘀 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗺𝗮𝗽𝘀: → Name it project-map, set its effort to medium, and preload a design skill → The first time, it asks what style you like and saves it to memory. After that, every map fits your taste and the project at hand → Call it before you let Opus run on its own. It draws the map in the background in a few minutes while the main session keeps coding on high, and it updates the map after every milestone → The map shows four things: the parts the project breaks into, how far each part has gotten, 𝘄𝗵𝗮𝘁'𝘀 𝘀𝘁𝘂𝗰𝗸 𝗮𝗻𝗱 𝘄𝗵𝗮𝘁 𝗶𝘁'𝘀 𝘄𝗮𝗶𝘁𝗶𝗻𝗴 𝗼𝗻, and what to do next (if you don't make the call, it goes ahead with that) Send this prompt to Opus 5.5 in Claude Code 👇 "Set up a subagent that only draws project maps: 1. Create project-map in ~/.claude/agents: model: opus, effort: medium, memory: user. Preload the design skills I have installed (for example impeccable). It needs to read the code, git history, and issues (if there are any). It only writes to a .project-map/ folder in the project, and adds that folder to .gitignore. 2. The first time, run it in the foreground so it can ask me what style I like: dark or light, and one accent color. It saves my answer to memory and follows it every time. If I already have dashboard-builder, reuse the style it saved and don't ask me again. 3. The map is one HTML file that opens with a double-click. It breaks the project into a few main parts and marks each one done, in progress, not started, or stuck. For stuck parts, say what they're waiting on. Milestones live in the map: if I can't name them, read the README and commit history and propose a first version for me to edit. At the top, show how many things are left before the next milestone and a suggested next step, and highlight the parts that changed since the lastview source ↗
-
• twitter 1w positiveThat’s quality 😂👏 https://t.co/HZRHHNKY0u That’s quality 😂👏 https://t.co/HZRHHNKY0uview source ↗
-
• twitter 1w positiveJump into an epic journey as Mickey Mouse through Wasteland, a realm of forgotten Disney characters. Will you become the epic hero Wasteland needs? Play Disney Epic Mickey: Rebrushed now on Nintendo Switch 2: https://t.co/ogwunCh5ok https://t.co/WaYeYkw4NC Jump into an epic journey as Mickey Mouse through Wasteland, a realm of forgotten Disney characters. Will you become the epic hero Wasteland needs? Play Disney Epic Mickey: Rebrushed now on Nintendo Switch 2: https://t.co/ogwunCh5ok https://t.co/WaYeYkw4NCview source ↗
-
• twitter 1w positiveEveryone is sick of generic AI, including us. Gamma was the first AI presentation platform to reach real scale. But now that AI tools are everywhere, everything looks the same. Earlier this summer, we came to an uncomfortable conclusion: without a step change in visual variety, Gamma’s output would look and feel too … Everyone is sick of generic AI, including us. Gamma was the first AI presentation platform to reach real scale. But now that AI tools are everywhere, everything looks the same. Earlier this summer, we came to an uncomfortable conclusion: without a step change in visual variety, Gamma’s output would look and feel too similar to all the other AI tools. The entire point of a presentation is to help someone "get" your idea, to see it as unique and do something with it. We started over. Today we're shipping Gamma 5, the biggest product overhaul in our company's history. We rebuilt Gamma to create presentations that actually look like you and your brand. Your teams can design anything they dream up, following your brand's aesthetic or something entirely new, with every frontier and image model under the hood. And the same goes for your docs, social assets, and graphics. We revamped our agent, design tools, editing, import, export, connectors, and more. Get your ideas out there.view source ↗
-
• twitter 1w positiveSpaceXAI engineer Lauren Tan: "99% of people use GrokBot for only a tiny fraction of what it can actually do. They run a single agent without using loops or graphs. I’m running a fully autonomous team of 20+ GrokBot agents. I have a Chief of Staff agent, a PM agent, and more than 20 specialized workers. This is what … SpaceXAI engineer Lauren Tan: "99% of people use GrokBot for only a tiny fraction of what it can actually do. They run a single agent without using loops or graphs. I’m running a fully autonomous team of 20+ GrokBot agents. I have a Chief of Staff agent, a PM agent, and more than 20 specialized workers. This is what the new engineering stack looks like. In a 1 hour session, a SpaceXAI engineer demonstrated how to build a powerful team of AI agents from the ground up. This is worth more than a $500 agentic engineering course. Watch this workshop today and learn how to build your own fleet of GrokBot agents.view source ↗
-
• twitter 1w positiveIf you want to work with me at @triggerdotdev, we're hiring backend engineers, a dev marketer, a devrel (SF) and a design engineer 😃 If you want to work with me at @triggerdotdev, we're hiring backend engineers, a dev marketer, a devrel (SF) and a design engineer 😃view source ↗
-
• twitter 1w positiveEverything Jason does is insane. I've been using it for two days trying to get work for this agency I'm cooking up. Bro, for $10 yesterday, I've already had 30 conversations in two days with people hiring. I'm telling you, this is probably the best tool I swear ive ever used. Everything Jason does is insane. I've been using it for two days trying to get work for this agency I'm cooking up. Bro, for $10 yesterday, I've already had 30 conversations in two days with people hiring. I'm telling you, this is probably the best tool I swear ive ever used.view source ↗
-
• twitter 1w positiveThis SEO strategy works pretty well. A real-world example: https://t.co/a1UfpjTlVW This SEO strategy works pretty well. A real-world example: https://t.co/a1UfpjTlVWview source ↗
-
• twitter 1w positiveDeepseek-v4.1-flash can build 3d worlds like no other local model, it's incredible at this kind of task. 1 shot! 40 score on AA for this model is just wrong, it's so good. It's just weak in UI, and has about 200M token window before it goes off track https://t.co/KmFOZso3YY Deepseek-v4.1-flash can build 3d worlds like no other local model, it's incredible at this kind of task. 1 shot! 40 score on AA for this model is just wrong, it's so good. It's just weak in UI, and has about 200M token window before it goes off track https://t.co/KmFOZso3YYview source ↗
-
• twitter 1w positiveI’m not really happy that my TikTok accounts are being exposed here either. I used to be fine with it because I genuinely enjoyed sharing. But well, game is game. A lot of people have already copied my content 1:1. Instead of getting mad at them, I’ve learned to accept that this is just how the market works. So this… I’m not really happy that my TikTok accounts are being exposed here either. I used to be fine with it because I genuinely enjoyed sharing. But well, game is game. A lot of people have already copied my content 1:1. Instead of getting mad at them, I’ve learned to accept that this is just how the market works. So this time, I’m going to scale 100x harder. 👨🌾👨🌾 Copycats don’t make me want to quit. They make me want to grow even faster.view source ↗