•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 626–650 of 1145 mentions
-
• twitter 1w negativeHarvard and Wharton researchers ran 776 Procter & Gamble professionals through real product work. alone or in pairs. with AI or without. one person with AI matched the work of a two-person team without it. this 2-page PDF breaks down "The Cybernetic Teammate" study, and the finding that matters most if you work alon… Harvard and Wharton researchers ran 776 Procter & Gamble professionals through real product work. alone or in pairs. with AI or without. one person with AI matched the work of a two-person team without it. this 2-page PDF breaks down "The Cybernetic Teammate" study, and the finding that matters most if you work alone: AI filled in the expertise the missing teammate would have brought. the real question isn't "can AI replace a team?" it's "which seat in my one-person company is still empty?" how to fill every seat is in the article below:view source ↗
-
• twitter 1w negativeNightmare fuel... 📺: THE STRANGERS https://t.co/Fmnm7GGbxf Nightmare fuel... 📺: THE STRANGERS https://t.co/Fmnm7GGbxfview source ↗
-
• twitter 1w negativeWe haven't stopped growing since we launched our new ChatGPT plugin last Tuesday. I've never seen anything like this, we added 500K in ARR this week only! https://t.co/XPy8wxdrEI We haven't stopped growing since we launched our new ChatGPT plugin last Tuesday. I've never seen anything like this, we added 500K in ARR this week only! https://t.co/XPy8wxdrEIview source ↗
-
• twitter 1w negativei'd judge a personal agent by what i can take with me when i leave. chat history isn't enough. i want the corrections that changed how it acts. portable memory cuts lock-in, but carrying every old rule creates a new problem. i'd review the rules before moving them. i'd judge a personal agent by what i can take with me when i leave. chat history isn't enough. i want the corrections that changed how it acts. portable memory cuts lock-in, but carrying every old rule creates a new problem. i'd review the rules before moving them.view source ↗
-
• twitter 1w negativemultiple people pointed out that this wording is more confusing than it needs to be. agreed. shipping an update to have this say "Add custom MCP server"! multiple people pointed out that this wording is more confusing than it needs to be. agreed. shipping an update to have this say "Add custom MCP server"!view source ↗
-
• twitter 1w negativeGPT-6.1 Sol could make Astra your backup for daily work. Test the cheaper option first. → Use Sol for coding, long documents, and tasks with several steps. → Start with lower reasoning effort. → Check the result against work you know well. → Turn effort up when needed. → Save Astra for the hardest science problem… GPT-6.1 Sol could make Astra your backup for daily work. Test the cheaper option first. → Use Sol for coding, long documents, and tasks with several steps. → Start with lower reasoning effort. → Check the result against work you know well. → Turn effort up when needed. → Save Astra for the hardest science problems. Use Luna for simple jobs at scale. OpenAI reports Sol’s standard token rates at roughly one-fifth of Astra’s. Save this video, you’ll know how to test a lighter model for your daily work. Want the SOP? DM me. 💬view source ↗
-
• twitter 1w negativeIf you run a cold email agency and you're stuck... I get it. I talk to agency owners about this all day. And the same thing keeps coming up: * You've got the tools * You've got the inboxes * You're scraping leads * You're running campaigns * You're booking some meetings * But it's not scaling * Clients are churning.… If you run a cold email agency and you're stuck... I get it. I talk to agency owners about this all day. And the same thing keeps coming up: * You've got the tools * You've got the inboxes * You're scraping leads * You're running campaigns * You're booking some meetings * But it's not scaling * Clients are churning... You're stuck in the loop. Here's the problem: it's your clients. Specifically - their offer sucks. And they don't have the balls to fix it. That's the game. You could fire 80% of your clients right now and make more money with the 20% that are left. And they'd get results. But you WON'T. Because when you've got rent to pay and someone's trying to hand you $3,000 a month... It doesn't matter that they're a recruiter trying to get into healthcare. You want the money, bro. And when there's strong demand - like a "strategic advisory firm looking for acquisitions" - you say yes. Even if they're asking for 100 meetings a month with founders ready to sell. You can't fulfill that. The reason there's such strong demand is because you can't fulfill that. But if a company's trying to pay you $10K a month to try? Who's gonna tell them no? That - at a smaller scale - is the reason your business isn't working. Here's an example: Had a Galadon Gold member come in last week. Said he wanted to offer "marketing consulting and lead gen to agencies" A younger version of me might've let him cook... Pat him on the back, collect the check, send some emails, watch him churn in 60 days. But now? I asked two questions: What's your background? What's the biggest ROI win you've delivered? Turns out, he'd spent the last 10 years in fintech. And the biggest result? He helped a fintech company go from zero leads at a trade show to 150 leads in three days - same booth, same spend, totally different execution. I'm like... that's the offer. Forget "marketing consulting" or "growth strategy" or "I do ads." "I help fintech companies turn trade shows fromview source ↗
-
• twitter 1w negativeWork on hard problems people want Work on hard problems people wantview source ↗
-
• twitter 1w negativeThe process to submit plugins for ChatGPT is much easier now. You can upload your zip, fix validation issues, then submit for review. When approved, you choose when to publish. See the process here: https://t.co/KLTQy1uLZq https://t.co/JkyHStLtLb The process to submit plugins for ChatGPT is much easier now. You can upload your zip, fix validation issues, then submit for review. When approved, you choose when to publish. See the process here: https://t.co/KLTQy1uLZq https://t.co/JkyHStLtLbview source ↗
-
• twitter 1w negativesome other stuff that i didn't like got fixed, and the render time is now halved. probably won't update this anymore, unless i receive some pr that magically makes everything more efficient https://t.co/ozz054z3d3 some other stuff that i didn't like got fixed, and the render time is now halved. probably won't update this anymore, unless i receive some pr that magically makes everything more efficient https://t.co/ozz054z3d3view source ↗
-
• twitter 1w negativeRough draft: recreated this with Wireflow using only existing clips. Sticker titles, character switcher, phone frame, all assembled by Claude + @wireflowai Not perfect yet. The skill to make these is open source: https://t.co/nVT7v60tUF https://t.co/f4VjcKvQv9 Rough draft: recreated this with Wireflow using only existing clips. Sticker titles, character switcher, phone frame, all assembled by Claude + @wireflowai Not perfect yet. The skill to make these is open source: https://t.co/nVT7v60tUF https://t.co/f4VjcKvQv9view source ↗
-
• twitter 1w negative@mitchellh @wolfie_ Very weird and aggro community. Not surprised they are coming after you. Disappointed, but not surprised. @mitchellh @wolfie_ Very weird and aggro community. Not surprised they are coming after you. Disappointed, but not surprised.view source ↗
-
• twitter 1w negativewhy do you work on hard problems? https://t.co/OKk1UyO2kK why do you work on hard problems? https://t.co/OKk1UyO2kKview source ↗
-
• twitter 1w negativethe only reason i haven't switched to grokbot or muse is because i want to own my agent... i do not like the idea of being locked into a companies ecosystem the only reason i haven't switched to grokbot or muse is because i want to own my agent... i do not like the idea of being locked into a companies ecosystemview source ↗
-
• twitter 1w negativeYeah I'm aware of the increaseCapacity issues and a had a fix for exactly those benchmarks but reverted it because I couldn't find a real world workload it helped and it had other adverse effects (real wold workloads that expanded styles a bit and then introduced a bunch of graphemes regressed). I prioritized the load … Yeah I'm aware of the increaseCapacity issues and a had a fix for exactly those benchmarks but reverted it because I couldn't find a real world workload it helped and it had other adverse effects (real wold workloads that expanded styles a bit and then introduced a bunch of graphemes regressed). I prioritized the load I had, for now. And yep! Expanded cells is also something already on the roadmap. We now have auto-compression for pages leaving the active area. The missing hook was there, but we have it now, so just a matter of time.view source ↗
-
• twitter 1w negativeIf a sales call starts with 20 minutes of explaining, your positioning is broken. Fxes: 1. Lead with the painful outcome 2. Delete feature-first copy 3. Name the wrong competitor 4. Give champions internal language 5. Make every team member use the same sentence If a sales call starts with 20 minutes of explaining, your positioning is broken. Fxes: 1. Lead with the painful outcome 2. Delete feature-first copy 3. Name the wrong competitor 4. Give champions internal language 5. Make every team member use the same sentenceview source ↗
-
• twitter 1w negativeWas on a customer call with someone who lived in a "retirement" community. She mentioned she's gifted our product to 10+ of her friends in the community and is always talking about it -- so I was curious to see if anyone near her ended up buying the product. Historically this would be...very hard to figure out. > … Was on a customer call with someone who lived in a "retirement" community. She mentioned she's gifted our product to 10+ of her friends in the community and is always talking about it -- so I was curious to see if anyone near her ended up buying the product. Historically this would be...very hard to figure out. > now: @shopify plugin to codex. > Ask chat to map every single order on a street level. Map shows me 4 people in her community that have bought the product. Overlay that with an exclusion from paid sources, and there's a fairly decent chance they became customers because of her. then asked chat to build filters to find out where are there unusually high levels of repeat + a number of other things i wanted to filter by. really crazy. There is a terrible amount that you can do now, that prevoiusly you couldn't imagine.view source ↗
-
• twitter 1w negativeIf I had to say in one word what distinguishes the juicy ones, the word might be unity. Nasty problems tend to be collections of random, often extrinsic constraints. Juicy problems have one main thing you're trying to solve If I had to say in one word what distinguishes the juicy ones, the word might be unity. Nasty problems tend to be collections of random, often extrinsic constraints. Juicy problems have one main thing you're trying to solveview source ↗
-
• twitter 1w negativeOne very important thing that even the smartest people seem to have to learn from experience is taste in problems. There are juicy problems that tend to lead to interesting discoveries, and nasty problems that tend not to. One very important thing that even the smartest people seem to have to learn from experience is taste in problems. There are juicy problems that tend to lead to interesting discoveries, and nasty problems that tend not to.view source ↗
-
• twitter 1w negativethe "Hard Problem" in a nutshell: 1. atoms aren't conscious 2. humans are made of atoms 3. humans are conscious 4. so at least one of 1-3 must be false Panpsychists say #1 is false Dualists say #2 is false Illusionists say #3 is false Materialists say #4 is false the "Hard Problem" in a nutshell: 1. atoms aren't conscious 2. humans are made of atoms 3. humans are conscious 4. so at least one of 1-3 must be false Panpsychists say #1 is false Dualists say #2 is false Illusionists say #3 is false Materialists say #4 is falseview source ↗
-
• twitter 1w negativeOmarchy QA update There are two things that are very hard about building a QA agent harness. 1. How to go fast in the day and age of agents (which are very very slow) 2. What is correct? What do we even test? The first one is less exciting but the second one has been this last weeks challenge. When I originally … Omarchy QA update There are two things that are very hard about building a QA agent harness. 1. How to go fast in the day and age of agents (which are very very slow) 2. What is correct? What do we even test? The first one is less exciting but the second one has been this last weeks challenge. When I originally started this, to have a basic agent crawl the desktop in QEMU was about ~15 minutes to open the lock screen and then type in password and confirm that we are back on the desktop Then a few changes I was down to ~5 for median. That sucks. With more changes and now a custom harness I am down to ~2 minutes. That still sucks. So for the last week I have been experimenting with using Clef (cloudflare's jev, it has image capabilities, which is what i needed) and I am hoping to get this number down from 2 minutes to 30 seconds for * logging in * system menu * lock screen * password * back to desktop * confirm i am done. If I can get this then I can officially move to step 2, which... well I am not stoked about step 2 and may need some help. So if you are interested in helping, let me know.view source ↗
-
• twitter 1w negativeOne very important thing that even the smartest people seem to have to learn from experience is taste in problems. There are juicy problems that tend to lead to interesting discoveries, and nasty problem that tend not to. One very important thing that even the smartest people seem to have to learn from experience is taste in problems. There are juicy problems that tend to lead to interesting discoveries, and nasty problem that tend not to.view source ↗
-
• twitter 1w negativereally against every single platform having a developer portal that you have to submit mcps through imagine how much the web would suck if every website you visited in chrome had to go through a review process with google first what are we doing!? really against every single platform having a developer portal that you have to submit mcps through imagine how much the web would suck if every website you visited in chrome had to go through a review process with google first what are we doing!?view source ↗
-
• twitter 1w negativeAgent routing can run on localhost now. Routing, tool selection and output checks are all classification calls. A decision model answers each one in a single forward pass with zero generated tokens, so stock llama-server handles them and every hop stops being a hosted API request. Pick the size per check: Julia-1 at … Agent routing can run on localhost now. Routing, tool selection and output checks are all classification calls. A decision model answers each one in a single forward pass with zero generated tokens, so stock llama-server handles them and every hop stops being a hosted API request. Pick the size per check: Julia-1 at 144M for cheap gates, Kev-4B as a 3 GB quant, OpenJev 27B when a wrong call is expensive.view source ↗
-
• twitter 1w negativeWas just randomly watching adipurush trailer. And look what I found ? 😭🤣🤣🤣🤣🤣🤣 https://t.co/UhiO6qABZi Was just randomly watching adipurush trailer. And look what I found ? 😭🤣🤣🤣🤣🤣🤣 https://t.co/UhiO6qABZiview source ↗