•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 976–1000 of 1145 mentions
-
• twitter 1w negativeCornell Young IV rides for Shaquille 😭 https://t.co/yD6tTqmqe1 Cornell Young IV rides for Shaquille 😭 https://t.co/yD6tTqmqe1view source ↗
-
• twitter 1w negative🚨 codex users!! this is what you are getting for $20 GPT-6.1 Sol at Max (using fast) > 37 Million tokens > 53 Minutes of work > 48% of your WEEKLY usage!!! > $18 API usage Its EXTREMELY SLOW on normal The worst part is you are only getting 300 Million tokens ONLY per MONTH!! PLEASE!! PLEASE DO NOT WASTE YOUR M… 🚨 codex users!! this is what you are getting for $20 GPT-6.1 Sol at Max (using fast) > 37 Million tokens > 53 Minutes of work > 48% of your WEEKLY usage!!! > $18 API usage Its EXTREMELY SLOW on normal The worst part is you are only getting 300 Million tokens ONLY per MONTH!! PLEASE!! PLEASE DO NOT WASTE YOUR MONEY!! NOTE: I don't have a 5h limit and speed is ONLY 12 tokens/sec on normal requests. Used max because of absolute best quality.view source ↗
-
• twitter 1w negative10,000,000 views, 4 day old account btw 😭 theres like 100 of this accounts all over my feed. https://t.co/XcbujbJQF1 10,000,000 views, 4 day old account btw 😭 theres like 100 of this accounts all over my feed. https://t.co/XcbujbJQF1view source ↗
-
• twitter 1w negativeHey @grok I'm genuinely Confused, Which part of Earth is this ? https://t.co/IMMNEFMHIc Hey @grok I'm genuinely Confused, Which part of Earth is this ? https://t.co/IMMNEFMHIcview source ↗
-
• twitter 1w negativeDo yourself a favor this month: - Pick one boring, painful problem people pay to fix - Build an AI agent that solves it end to end - Give it real tools, memory and MCP connectors - Add guardrails: approvals, limits, error handling - Write evals so you can prove it works - Track cost and speed for every run - De… Do yourself a favor this month: - Pick one boring, painful problem people pay to fix - Build an AI agent that solves it end to end - Give it real tools, memory and MCP connectors - Add guardrails: approvals, limits, error handling - Write evals so you can prove it works - Track cost and speed for every run - Deploy it with a live link anyone can try - Open-source the code on GitHub with a clean README - Record a 60-second demo video - Write a short post on what broke and how you fixed it - Get 10 real people to use it - Put the link, demo and numbers on your resume Recruiters skim degrees. They stop scrolling for a live link. Bookmark this + send it to someone still "learning AI" with nothing to show.view source ↗
-
• twitter 1w negativeMajor green flags in an AI builder: 1/ ships before they feel ready 2/ has a live link, not just a GitHub repo 3/ reads the docs before asking Twitter 4/ knows when AI is the wrong answer 5/ writes evals not just prompts 6/ can explain what their agent does in one sentence 7/ tracks cost per task without being asked 8… Major green flags in an AI builder: 1/ ships before they feel ready 2/ has a live link, not just a GitHub repo 3/ reads the docs before asking Twitter 4/ knows when AI is the wrong answer 5/ writes evals not just prompts 6/ can explain what their agent does in one sentence 7/ tracks cost per task without being asked 8/ breaks their own product before users do 9/ learns a new tool every month, masters one every year 10/ posts what broke not just what worked 11/ has 10 real users before talking about scale 12/ doesn't switch stacks every time a new model drops 13/ helps other builders without expecting anything back 14/ finishes side projects instead of starting new ones 15/ can still code when the AI is down If you hit most of these, drop what you're building 👇view source ↗
-
• twitter 1w negativeAre they couple😭😭 https://t.co/ORN43JqEUR Are they couple😭😭 https://t.co/ORN43JqEURview source ↗
-
• twitter 1w negativeI am very uncomfortable about people trying to ascribe religious force or a surrender of human judgment to AI models, and think it is a real safety issue. I am very uncomfortable about people trying to ascribe religious force or a surrender of human judgment to AI models, and think it is a real safety issue.view source ↗
-
• twitter 1w negativebro can eat Tourist in his breakfast 😭😭 10/10 questions of ICPC solved by bro without even solving CF, CC questions, bro is a LC guy https://t.co/sWCESI0txr bro can eat Tourist in his breakfast 😭😭 10/10 questions of ICPC solved by bro without even solving CF, CC questions, bro is a LC guy https://t.co/sWCESI0txrview source ↗
-
• twitter 1w negativeFolks iv been mentally going through a lot and the problem is a sheer lack of focus for past ~6 months or so, hence the struggle making videos. docs/psychs havent worked though i continue to search better ones. but im excited this month as gareeb Scientist completes 10 years. I have some 8/10 vids in backlog i want to… Folks iv been mentally going through a lot and the problem is a sheer lack of focus for past ~6 months or so, hence the struggle making videos. docs/psychs havent worked though i continue to search better ones. but im excited this month as gareeb Scientist completes 10 years. I have some 8/10 vids in backlog i want to get done with., tbh i know ill be struggling to get maybe even 2 done, i just want you to know im trying...view source ↗
-
• twitter 1w negativeOpus 5.5 gave me a huge problem guys. It finished all my work and I don't know what to do next. Opus 5.5 gave me a huge problem guys. It finished all my work and I don't know what to do next.view source ↗
-
• twitter 1w negativewe are all bad in someone’s story. we are all bad in someone’s story.view source ↗
-
• twitter 1w negativePaying an agency $5,000 a month doesn't mean your content is done right. Watched a founder pull up a post his agency shipped minutes earlier. Broken tags. Four posts a month. Five grand. The bar in this industry is on the floor. Paying an agency $5,000 a month doesn't mean your content is done right. Watched a founder pull up a post his agency shipped minutes earlier. Broken tags. Four posts a month. Five grand. The bar in this industry is on the floor.view source ↗
-
• twitter 1w negativeChrista Pike remains critically ill after two failed execution attempts, renewing scrutiny of Tennessee's lethal injection process and sparking debate over the death penalty https://t.co/B9j0vxOzcn https://t.co/PdDsDodEPx Christa Pike remains critically ill after two failed execution attempts, renewing scrutiny of Tennessee's lethal injection process and sparking debate over the death penalty https://t.co/B9j0vxOzcn https://t.co/PdDsDodEPxview source ↗
-
• twitter 1w negativewhat are your biggest issues with OpenCode 2? what are your biggest issues with OpenCode 2?view source ↗
-
• twitter 1w negativeWhat’s the ONE thing Hermes Agent is missing? If you also use Codex, Claude Code, or another agent, what task makes you switch away from Hermes? • A feature you can’t find? • A workflow that takes fewer steps elsewhere? • A task another agent handles more reliably? Finish this sentence:✋ “I’d use Hermes more if it … What’s the ONE thing Hermes Agent is missing? If you also use Codex, Claude Code, or another agent, what task makes you switch away from Hermes? • A feature you can’t find? • A workflow that takes fewer steps elsewhere? • A task another agent handles more reliably? Finish this sentence:✋ “I’d use Hermes more if it could ______.” Name one feature + the real task it would help you finish. 👇view source ↗
-
• twitter 1w negativeToday when I turned on Cricbuzz to see the score between India and West Indies, see what happened. When I turned on the commentary, the 33Over to 48.6 Over ball was missing, meaning there was no commentary in the middle overs. Can anyone tell me what this is for? If you don't believe me, then watch the screen reco… Today when I turned on Cricbuzz to see the score between India and West Indies, see what happened. When I turned on the commentary, the 33Over to 48.6 Over ball was missing, meaning there was no commentary in the middle overs. Can anyone tell me what this is for? If you don't believe me, then watch the screen recording.view source ↗
-
• twitter 1w negativeI think some of our deepest regrets aren't about the things we did. They're about the versions of ourselves we abandoned because we were afraid of disappointing other people. How many of your life decisions were actually yours? I think some of our deepest regrets aren't about the things we did. They're about the versions of ourselves we abandoned because we were afraid of disappointing other people. How many of your life decisions were actually yours?view source ↗
-
• twitter 1w negativeThese unsexy problems are goldmines These unsexy problems are goldminesview source ↗
-
• twitter 1w negative🚨 BREAKING 🚨 WEEKEND KA VAAR UPDATE Ayushmann Khurrana asked Qazi if he has ever actually done MMA. #QaziTouqeer : NO. 😭 Ayushmann: Then how did you decide you want to fight and BEAT KHABIB? 💀 Qazi’s explanation? Apparently Khabib showed up in his dream with two horses, beat him up badly, tore his clothes and lef… 🚨 BREAKING 🚨 WEEKEND KA VAAR UPDATE Ayushmann Khurrana asked Qazi if he has ever actually done MMA. #QaziTouqeer : NO. 😭 Ayushmann: Then how did you decide you want to fight and BEAT KHABIB? 💀 Qazi’s explanation? Apparently Khabib showed up in his dream with two horses, beat him up badly, tore his clothes and left him traumatised. 😭😭 So now Qazi wants revenge for a fight that literally happened IN HIS DREAM. Ayushmann basically said: “Bro, keep it a dream. In real life, Khabib will humble you for real.” 💀 LATEST NEWS: Qazi Is A Buddha #ScoutOp #BiggBoss20view source ↗
-
• twitter 1w negativeHopefully it is going to be the last meeting with my lawyer on this stupid case. Looking back, that 1 lakh of security deposit (my share on the amount) was absolutely negligible compared to the amount of trauma that came from trying to retrieve it. It costed me more money and also so much time. Also had to deal with t… Hopefully it is going to be the last meeting with my lawyer on this stupid case. Looking back, that 1 lakh of security deposit (my share on the amount) was absolutely negligible compared to the amount of trauma that came from trying to retrieve it. It costed me more money and also so much time. Also had to deal with the worst kinds of humans in this process. One of the flatmates decided to give up their share of the amount and chose peace over all this fighting and I think they made the right decision. It is important to analyse the second order effects of decisions. 1 lakh in hindsight was nothing and I should have just left it.view source ↗
-
• twitter 1w negativeBREAKING: The U.S. Coast Guard has launched a search for a missing Boston-bound plane off the coast of Nantucket after the FAA lost contact with the aircraft carrying six people. Get the latest updates with the Fox News app. Download here: https://t.co/Q8qTIP81ow https://t.co/OXds6NhCDP BREAKING: The U.S. Coast Guard has launched a search for a missing Boston-bound plane off the coast of Nantucket after the FAA lost contact with the aircraft carrying six people. Get the latest updates with the Fox News app. Download here: https://t.co/Q8qTIP81ow https://t.co/OXds6NhCDPview source ↗
-
• twitter 1w negativeIf I had 6 months to become an AI Evals Engineer. I’d do this. Stage 1: Testing Foundations and Statistics for AI - Learn: pytest (fixtures, parametrize), JSONL pipelines, precision/recall/F1, confidence intervals, Cohen's kappa. - Practice: Build a pytest plugin that loads a JSONL eval dataset and emits a metric r… If I had 6 months to become an AI Evals Engineer. I’d do this. Stage 1: Testing Foundations and Statistics for AI - Learn: pytest (fixtures, parametrize), JSONL pipelines, precision/recall/F1, confidence intervals, Cohen's kappa. - Practice: Build a pytest plugin that loads a JSONL eval dataset and emits a metric report on every run. - Why: Evals are tests and tests are math. If you cannot quantify agreement, you cannot prove improvement. Stage 2: LLM Failure Modes and Behavior - Learn: temperature and top-p sampling, seed control, tokenization edge cases, instruction drift, hallucination taxonomies. - Practice: Fuzz one model with 500 adversarial inputs and classify every failure into a taxonomy with reproduction steps. - Why: Every eval suite is a map of known failure modes. You cannot evaluate what you do not understand. Stage 3: Golden Dataset Engineering - Learn: annotation guidelines, stratified sampling, inter-annotator agreement, dataset versioning, contamination detection. - Practice: Build a 300-case golden dataset with 2 annotators, measure kappa, and version it in git with a changelog. - Why: Your evals are only as good as your dataset. Garbage labels produce garbage confidence. Stage 4: Scoring Methods and LLM-as-a-Judge - Learn: exact and fuzzy match, embedding similarity, rubric design, judge model selection, judge biases (verbosity, position, self-preference). - Practice: Build an LLM-as-a-judge with a rubric, calibrate it against 100 human labels and publish the agreement score. - Why: Judges drift and lie. Calibrated judges are instruments; uncalibrated ones are vibes. Stage 5: RAG Evaluation - Learn: hit rate, MRR, NDCG, recall@k, faithfulness vs relevance, citation grounding, abstention quality. - Practice: Build a RAG eval harness with separate retrieval and generation gates, plus adversarial queries that must return "no answer". - Why: RAG fails in two places. A single score hides which one is broken. Stage 6: Ageview source ↗
-
• twitter 1w negativeNo parent wants to have a broke son, no woman who wants to marry a broke man, no son who wants a broke father, no sister wants a broke brother. A man without money has no voice. As a man, fix your finances before you try to fix the world. No parent wants to have a broke son, no woman who wants to marry a broke man, no son who wants a broke father, no sister wants a broke brother. A man without money has no voice. As a man, fix your finances before you try to fix the world.view source ↗
-
• twitter 1w negative5) The slow week "Here are my last 3 months of sales: [paste]. Which 10 past customers should I contact this week, and what should I say to each one?" 5) The slow week "Here are my last 3 months of sales: [paste]. Which 10 past customers should I contact this week, and what should I say to each one?"view source ↗