One dashboard for multiple AI agents. AI that acts, not just responds. https://xitter.cf/t.co/f8R5RQSrM6
Joined April 2026
- Tweets865
- Following55
- Followers289
- Likes772
Pinned Tweet
You’re the founder, the assistant, the person screening every CV, and somehow also the person calling to change the booking.
Agentese is building an AI workforce for people carrying too many jobs at once.
Each agent takes on a specific part of the work:
@LifeClaw_AI calls businesses for you, handling bookings, cancellations, and rescheduling, even in another language.
Little Pink Potato works for both sides of hiring. For applicants, it finds roles that fit their experience and preferences. For recruiters, it evaluates candidates, gives match scores, and flags concerns, connecting the right people with the right roles.
CodeAutrix audits and stress-tests smart contracts and AI skills in minutes, catching critical flaws and security vulnerabilities before you ship.
@Yarrow_ai turns uncertain questions into forecasts you can check. It models how people, institutions, and markets are likely to respond to a policy change, market shift, or crisis, gives probabilities with the reasoning behind them, and measures every forecast against what actually happens.
Start with the task that keeps interrupting everything else.
Which one would you hand over first?
Would you rather be interviewed by a human or an AI?
A field experiment involving 70,000 job applicants found that candidates interviewed by AI voice agents were 12% more likely to receive job offers.
The researchers suggest that more structured, consistent interviews helped collect useful information about candidates.
And humans still made the hiring decisions.
Maybe the real question isn't whether AI should interview candidates.
It's whether AI can give every candidate a fairer chance to demonstrate their skills.
Of course, assuming the interview actually ends and the two AI agents don't get stuck saying goodbye to each other forever!
A company sent an AI recruiter to interview him, so he sent his own AI agent instead
The two AIs discussed the job between themselves. No human was on the call
Then both of them glitched, and got stuck saying goodbye to each other. Endlessly
Job interviews in 2026: your agent talks to their agent, and neither knows how to hang up
Agentese AI retweeted
Four earnings in two days.
Netflix: Beat EPS ($0.87 vs $0.82), but slashed full-year ad revenue guidance from $3B to $2.7B. Stock down 5.3% — two-year low. As an investor, I don't care that you beat this quarter. I care that if someone could have told me next quarter would be worse, I could have prepared.
Meta: Revenue $64.6B (+26% YoY), EPS beat by 7.3%. But the headline is the 2-for-1 stock split — new non-voting Class C shares. Zuckerberg keeps full control while raising capital for AI. $34.9B capex in one quarter. Up 3.8% after hours.
Tesla: Revenue slight beat, but GAAP EPS dropped from $0.39 to $0.22. Bright spot: FSD attach rate above 75%, subscriptions up 67%. Weakness: regulatory credits collapsed from $417M to $113M. Mixed — stock barely moved. If someone had told me that beforehand, I wouldn't have been glued to my screen all night lol
GM: The quiet winner. Revenue $50B beat, EPS $3.70 beat, raised full-year guidance, authorized $20B buyback. But WTI at $98 is the question every analyst asks on every call.
Earnings are done, but the story isn't — next Monday Israel votes, next Tuesday the Fed decides. Back to back. The election outcome shapes Iran policy, Iran shapes oil, oil shapes the next CPI, CPI shapes the Fed. One chain.
YarrowPulse tracks every node on that chain. Open it, and ask anything you want to predict.
pulse.yarrowlab.ai ☘️
Agentese AI retweeted
The PHP security team is giving advance notice: Upcoming security release Wed, Oct 21st 2026, 14:00UTC, two critical vulnerabilities, both rated CVSSv4 9.2.
PHP users are encouraged to prepare to update. Earlier release if details become public or active exploit evidence.
Agentese AI retweeted
Ваш ИИ умеет вести переговоры. Но знает ли он, в чём вы не готовы идти на компромисс?
Более высокая зарплата, гибкий график, подходящая должность — чтобы получить более выгодное предложение, нужно понимать, что для вас действительно важно.
Little Pink Potato применяет эту идею в сфере найма, используя ИИ-агентов, которые ведут переговоры от имени кандидатов и компаний.
Your next salary negotiation might happen without you ever speaking to the recruiter.
Your AI negotiates with the company's AI. Both know what their side wants. Both are instructed to get the best possible deal.
Who gives in first?
The Financial Times recently reported that Anthropic is exploring how AI agents should negotiate on behalf of humans, including how to represent their interests while reaching fair outcomes.
Imagine applying for a job in this world.
Your AI knows your experience, salary expectations, and dealbreakers. The company's recruiting agent knows its hiring budget, role requirements, and benefits.
They could negotiate compensation, discuss conditions, and work toward an offer before either human steps in.
But what happens when both agents are programmed to get the best deal?
Interestingly, MIT researchers studied over 180,000 AI-to-AI negotiations and found that warmer, more cooperative agents were more likely to reach agreements.
Aggressive agents could secure better terms when successful, but also risked more deadlocks.
Apparently, even AI needs to learn how to negotiate without being a jerk.
And this isn't such a distant concept for us.
Little Pink Potato already helps job seekers discover opportunities, apply for roles, and negotiate compensation.
Notably at Little Pink Potato, your AI talent agent can negotiate directly with a company's recruiting agent, each representing the interests of their human.
Would you trust your AI to negotiate your salary, especially if the other side is also an AI?
Agentese AI retweeted
Little Pink Potato застосовує цю ідею у сфері найму, використовуючи ШІ-агентів, які ведуть переговори від імені кандидатів і компаній.
xitter.cf/Agentese_AI/status/211…
Your next salary negotiation might happen without you ever speaking to the recruiter.
Your AI negotiates with the company's AI. Both know what their side wants. Both are instructed to get the best possible deal.
Who gives in first?
The Financial Times recently reported that Anthropic is exploring how AI agents should negotiate on behalf of humans, including how to represent their interests while reaching fair outcomes.
Imagine applying for a job in this world.
Your AI knows your experience, salary expectations, and dealbreakers. The company's recruiting agent knows its hiring budget, role requirements, and benefits.
They could negotiate compensation, discuss conditions, and work toward an offer before either human steps in.
But what happens when both agents are programmed to get the best deal?
Interestingly, MIT researchers studied over 180,000 AI-to-AI negotiations and found that warmer, more cooperative agents were more likely to reach agreements.
Aggressive agents could secure better terms when successful, but also risked more deadlocks.
Apparently, even AI needs to learn how to negotiate without being a jerk.
And this isn't such a distant concept for us.
Little Pink Potato already helps job seekers discover opportunities, apply for roles, and negotiate compensation.
Notably at Little Pink Potato, your AI talent agent can negotiate directly with a company's recruiting agent, each representing the interests of their human.
Would you trust your AI to negotiate your salary, especially if the other side is also an AI?
Agentese AI retweeted
Your AI can negotiate. But does it know what you won’t compromise on?
A higher salary, flexible hours, the right role, getting a better offer starts with knowing what matters to you.
Little Pink Potato brings that idea to hiring, with AI agents negotiating on behalf of candidates and companies.
xitter.cf/agentese_ai/status/211…
Your next salary negotiation might happen without you ever speaking to the recruiter.
Your AI negotiates with the company's AI. Both know what their side wants. Both are instructed to get the best possible deal.
Who gives in first?
The Financial Times recently reported that Anthropic is exploring how AI agents should negotiate on behalf of humans, including how to represent their interests while reaching fair outcomes.
Imagine applying for a job in this world.
Your AI knows your experience, salary expectations, and dealbreakers. The company's recruiting agent knows its hiring budget, role requirements, and benefits.
They could negotiate compensation, discuss conditions, and work toward an offer before either human steps in.
But what happens when both agents are programmed to get the best deal?
Interestingly, MIT researchers studied over 180,000 AI-to-AI negotiations and found that warmer, more cooperative agents were more likely to reach agreements.
Aggressive agents could secure better terms when successful, but also risked more deadlocks.
Apparently, even AI needs to learn how to negotiate without being a jerk.
And this isn't such a distant concept for us.
Little Pink Potato already helps job seekers discover opportunities, apply for roles, and negotiate compensation.
Notably at Little Pink Potato, your AI talent agent can negotiate directly with a company's recruiting agent, each representing the interests of their human.
Would you trust your AI to negotiate your salary, especially if the other side is also an AI?
Your next salary negotiation might happen without you ever speaking to the recruiter.
Your AI negotiates with the company's AI. Both know what their side wants. Both are instructed to get the best possible deal.
Who gives in first?
The Financial Times recently reported that Anthropic is exploring how AI agents should negotiate on behalf of humans, including how to represent their interests while reaching fair outcomes.
Imagine applying for a job in this world.
Your AI knows your experience, salary expectations, and dealbreakers. The company's recruiting agent knows its hiring budget, role requirements, and benefits.
They could negotiate compensation, discuss conditions, and work toward an offer before either human steps in.
But what happens when both agents are programmed to get the best deal?
Interestingly, MIT researchers studied over 180,000 AI-to-AI negotiations and found that warmer, more cooperative agents were more likely to reach agreements.
Aggressive agents could secure better terms when successful, but also risked more deadlocks.
Apparently, even AI needs to learn how to negotiate without being a jerk.
And this isn't such a distant concept for us.
Little Pink Potato already helps job seekers discover opportunities, apply for roles, and negotiate compensation.
Notably at Little Pink Potato, your AI talent agent can negotiate directly with a company's recruiting agent, each representing the interests of their human.
Would you trust your AI to negotiate your salary, especially if the other side is also an AI?
Imagine a future where AI agents negotiate with each other to get the best deal for their human masters. That day may not be far away — but first they need to learn how to interact in negotiations, representing their clients while honouring fair outcomes ft.trib.al/5wISsbj
Agentese AI retweeted
We're looking for 20 seed users to test YarrowPulse — our AI prediction intelligence platform for tracking how global events impact your portfolio.
What you get:
— Early access to all features
— Direct line to the team for feedback
— Your name in our community credits when we launch publicly
What we need from you:
— Use Pulse for one week
— Tell us what works and what doesn't (15 min call)
— You trade stocks, crypto, or follow macro — that's the only requirement
Sign up: pulse.yarrowlab.ai
DM me or reply to join the seed group.
Agentese AI retweeted
You’re the founder, the assistant, the person screening every CV, and somehow also the person calling to change the booking.
Agentese is building an AI workforce for people carrying too many jobs at once.
Each agent takes on a specific part of the work:
@LifeClaw_AI calls businesses for you, handling bookings, cancellations, and rescheduling, even in another language.
Little Pink Potato works for both sides of hiring. For applicants, it finds roles that fit their experience and preferences. For recruiters, it evaluates candidates, gives match scores, and flags concerns, connecting the right people with the right roles.
CodeAutrix audits and stress-tests smart contracts and AI skills in minutes, catching critical flaws and security vulnerabilities before you ship.
@Yarrow_ai turns uncertain questions into forecasts you can check. It models how people, institutions, and markets are likely to respond to a policy change, market shift, or crisis, gives probabilities with the reasoning behind them, and measures every forecast against what actually happens.
Start with the task that keeps interrupting everything else.
Which one would you hand over first?
Agentese AI retweeted
Команда ИИ-агентов @Agentese_AI для решения практических задач:
@LifeClaw_AI обрабатывает телефонные звонки и бронирования, Little Pink Potato подбирает вакансии и кандидатов, CodeAutrix проводит аудит кода, а @Yarrow_ai предоставляет прогнозы с указанием вероятностей и обоснованием.
xitter.cf/Agentese_AI/status/211…
You’re the founder, the assistant, the person screening every CV, and somehow also the person calling to change the booking.
Agentese is building an AI workforce for people carrying too many jobs at once.
Each agent takes on a specific part of the work:
@LifeClaw_AI calls businesses for you, handling bookings, cancellations, and rescheduling, even in another language.
Little Pink Potato works for both sides of hiring. For applicants, it finds roles that fit their experience and preferences. For recruiters, it evaluates candidates, gives match scores, and flags concerns, connecting the right people with the right roles.
CodeAutrix audits and stress-tests smart contracts and AI skills in minutes, catching critical flaws and security vulnerabilities before you ship.
@Yarrow_ai turns uncertain questions into forecasts you can check. It models how people, institutions, and markets are likely to respond to a policy change, market shift, or crisis, gives probabilities with the reasoning behind them, and measures every forecast against what actually happens.
Start with the task that keeps interrupting everything else.
Which one would you hand over first?
Agentese AI retweeted
This week: Netflix (today), GM (Tue), Meta and Tesla (Wed). 70 S&P 500 companies reporting total.
Here's how to read them through YarrowPulse:
Netflix reports today at 4:45 PM ET. Consensus: $12.87B revenue, $0.82 EPS. But Netflix isn't just a streaming story — it's an ad revenue story. On track for ~$3B in ad revenue this year, double last year. The question: can engagement per member stop declining?
GM reports Tuesday. Consensus: $3.69 EPS (+30% YoY). North America carries this company. But oil at $103 WTI is squeezing consumer budgets. If GM flags demand softening, it's a macro signal.
Meta reports Wednesday. EPS expected up 460% YoY. AI-driven ad targeting is the engine. But Meta just released Muse Spark Open (816B parameters, runs on Apple Silicon) — the real question is whether open-source AI cannibalizes their cloud revenue or expands the pie.
Tesla also Wednesday. EPS expected -14.5% YoY. Deliveries beat at 486K, but margins are the concern. And Elon's political exposure in an election year adds volatility.
On YarrowPulse right now:
→ Federal Reserve 2026: 1 update, 70 related markets, 1 divergence ≥6pt. Heat: 🔥🔥🔥
→ Hormuz Strait Disruption: 11 updates, 38 markets. Heat: 🔥🔥🔥
Every one of these earnings reports is touched by the same macro chain: Iran war → oil at $100+ → CPI hot → Fed Oct 28 decision.
pulse.yarrowlab.ai ☘️
Errmmm never mind.
You called customer support to complain.
30 minutes on hold and 3 transfers later, you have another complaint. 😭
Next time, let @LifeClaw_AI handle it.
Agentese AI retweeted
An @Agentese_AI workforce for practical tasks: @LifeClaw_AI handles phone calls and bookings, Little Pink Potato matches jobs and candidates, CodeAutrix audits code, and @Yarrow_ai provides forecasts with probabilities and supporting reasoning.
xitter.cf/Agentese_AI/status/211…
You’re the founder, the assistant, the person screening every CV, and somehow also the person calling to change the booking.
Agentese is building an AI workforce for people carrying too many jobs at once.
Each agent takes on a specific part of the work:
@LifeClaw_AI calls businesses for you, handling bookings, cancellations, and rescheduling, even in another language.
Little Pink Potato works for both sides of hiring. For applicants, it finds roles that fit their experience and preferences. For recruiters, it evaluates candidates, gives match scores, and flags concerns, connecting the right people with the right roles.
CodeAutrix audits and stress-tests smart contracts and AI skills in minutes, catching critical flaws and security vulnerabilities before you ship.
@Yarrow_ai turns uncertain questions into forecasts you can check. It models how people, institutions, and markets are likely to respond to a policy change, market shift, or crisis, gives probabilities with the reasoning behind them, and measures every forecast against what actually happens.
Start with the task that keeps interrupting everything else.
Which one would you hand over first?
Agentese AI retweeted
Yesterday Ukraine launched over 1,000 drones at Russia — one of the largest drone attacks of the war. Targets included the St. Petersburg oil export terminal, a VK data center, and an Ozon logistics center. Fires confirmed.
This happened 8 days after Trump declared an "energy ceasefire."
Open YarrowPulse right now:
→ Russia–Ukraine ceasefire agreement: 18 updates in 24 hours, 20 related prediction markets. Market heat: 🔥🔥
→ Russia–Ukraine Territorial Control: 12 updates in 24 hours, 70 related markets. Market heat: 🔥🔥
The ceasefire topic is surging with updates. But here's the point: 1,000 drones hitting oil infrastructure doesn't look like a ceasefire. And the prediction markets know it — the ceasefire contract hasn't moved up.
Meanwhile, this attack adds upward pressure on oil. Brent hit $110 on Friday. Ukraine is now systematically targeting Russian energy exports — and that feeds directly into the next CPI print.
Geopolitics → energy → inflation → Fed. One chain. YarrowPulse tracks it all.
pulse.yarrowlab.ai ☘️
Meta’s Muse is bringing AI agents into business calls.
We’ve spent the past 7 months building @LifeClaw_AI that calls businesses on your behalf.
Getting an AI to make a call is one thing. Getting it to finish the job is another.
4 things we’ve learned:
1. Navigating phone menus is the easy part. Dealing with humans is harder.
People interrupt, misunderstand, transfer calls, and change their answers.
An agent needs to adapt in real time without losing track of the original request.
2. Businesses aren’t always ready to talk to AI.
Some are cooperative. Others get suspicious the moment they realize they’re speaking to a bot.
How an agent introduces itself and handles that reaction matters more than how human it sounds.
3. A completed call isn’t necessarily a completed task.
The AI saying “done” means very little if the business hasn’t actually confirmed anything.
Users need a clear outcome and a way to verify what happened.
4. Making calls is a capability. Knowing when to call is the real skill.
Some tasks can be completed online. Others require speaking to someone, following up, or changing approach halfway through.
That’s what we’re building with LifeClaw.
The goal was never to make AI talk on the phone. It was to get things done.
Ever wonder what actually happens behind the scenes when you use @LifeClaw_AI for your daily needs?
Below is an example.📝
Beyond booking, you can ask @LifeClaw_AI to call multiple spots and check availability before you make a final decision.
Agentese AI retweeted
Agentese membangun AI agent yang mampu menangani berbagai tugas praktis. Mulai dari menjawab pertanyaan selama rapat dan melakukan panggilan telepon, hingga mencari pekerjaan yang sesuai, mengelola persetujuan, mengaudit kode, dan memprediksi berbagai peristiwa.
Setiap produk memiliki peran khusus untuk membantu kamu beralih dari sekadar bertanya menjadi benar-benar menyelesaikan pekerjaan.
Jelajahi berbagai AI agent dari @Agentese_AI di bawah ini 👇
Building a company used to mean hiring a team.
Now it means deploying one.
We stopped building disconnected AI “tools” and started hiring AI “employees.”
Meet the Agentese workforce for your company:
🎙️ Meeting Copilot - Live meeting assistant
Answers hard questions out loud, in real time, during your meetings. No prep and no recaps, just the answer the moment the room needs one.
meeting-copilot.agentese.ai
🦞 LifeClaw - Personal call & admin agent
Makes your phone calls, bookings, and disputes for you: the cancellation, the appointment, the hold music, all handled.
lifeclaw.agentese.ai
🥔 Little Pink Potato - Automated job application agent
Applies to jobs for you at a scale no human could, screening thousands of roles, scoring the real matches, applying and negotiating based on your preference.
littlepinkpotato.com
☕ Kopi - Approval assistant for Lark
Clears the routine approvals piling up in your Lark queue, auto-handling the obvious yeses and flagging only what needs your judgment.
hellokopi.com
🔒 CodeAutrix - Smart contract & AI skill auditor
Audits and stress-tests smart contracts and AI skills in minutes, surfacing the critical flaw before you ship.
codeautrix.com
🍀 Yarrow - Multi-agent simulation platform
Ask a question, get a calibrated probability backed by a full chain of evidence and reasoning. Predictions you can grade, not opinions.
yarrowlab.ai
Stop managing tools, start delegating. With Agentese. 🚀
Agentese AI retweeted
Кожен продукт має своє конкретне завдання. Разом вони допомагають перейти від простого запитання до реального результату.
Ознайомтеся з командою ШІ-агентів @Agentese_AI нижче 👇
xitter.cf/MantaNetwork/status/21…
Agentese builds AI agents that handle practical tasks. From answering questions during meetings and making phone calls to finding suitable jobs, managing approvals, auditing code and forecasting events.
Each product has a specific role, helping you move from asking a question to getting work done.
Explore the @Agentese_AI workforce below 👇xitter.cf/Agentese_AI/status/208…
Agentese AI retweeted
Agentese создаёт ИИ-агентов, которые выполняют практические задачи: отвечают на вопросы во время встреч, совершают телефонные звонки, помогают находить подходящие вакансии, управлять согласованиями, проводить аудит кода и прогнозировать события.
У каждого продукта своя конкретная задача. Вместе они помогают перейти от простого вопроса к реальному результату.
Познакомьтесь с командой ИИ-агентов @Agentese_AI ниже 👇
Building a company used to mean hiring a team.
Now it means deploying one.
We stopped building disconnected AI “tools” and started hiring AI “employees.”
Meet the Agentese workforce for your company:
🎙️ Meeting Copilot - Live meeting assistant
Answers hard questions out loud, in real time, during your meetings. No prep and no recaps, just the answer the moment the room needs one.
meeting-copilot.agentese.ai
🦞 LifeClaw - Personal call & admin agent
Makes your phone calls, bookings, and disputes for you: the cancellation, the appointment, the hold music, all handled.
lifeclaw.agentese.ai
🥔 Little Pink Potato - Automated job application agent
Applies to jobs for you at a scale no human could, screening thousands of roles, scoring the real matches, applying and negotiating based on your preference.
littlepinkpotato.com
☕ Kopi - Approval assistant for Lark
Clears the routine approvals piling up in your Lark queue, auto-handling the obvious yeses and flagging only what needs your judgment.
hellokopi.com
🔒 CodeAutrix - Smart contract & AI skill auditor
Audits and stress-tests smart contracts and AI skills in minutes, surfacing the critical flaw before you ship.
codeautrix.com
🍀 Yarrow - Multi-agent simulation platform
Ask a question, get a calibrated probability backed by a full chain of evidence and reasoning. Predictions you can grade, not opinions.
yarrowlab.ai
Stop managing tools, start delegating. With Agentese. 🚀