The latest viral LinkedIn Posts

Aprendi isto mais tarde do que devia.

Às vezes o maior cuidado que podemos ter connosco não é terapia, nem meditação, nem mais sono.
É uma escolha. Sobre quem deixamos entrar. ❤️

Bom fim-de-semana.

https://lnkd.in/e7pRYg7C

Najgorsze jest to, że wielu menedżerów naprawdę wierzy, że robi dobrą robotę.

Kontrolują każdy szczegół.
Odpowiadają na maile w nocy.
„Ustawiają” ludzi zamiast ich rozwijać.

I jeszcze są z tego dumni.

Tylko że z boku to wygląda inaczej:
brak zaufania, zabita inicjatywa i zespół, który robi minimum, żeby przetrwać.

Dobry menedżer nie musi być wszędzie.
Nie musi wszystkiego kontrolować.
Nie musi udowadniać, że jest najważniejszy.

Buduje środowisko, w którym ludziom chce się pracować i brać odpowiedzialność.

To ta kluczowa różnica.

A Ty co o tym myślisz? Jakie masz z tym doświadczenia?

——

Zaobserwuj mój profil Tomasz Osman, żeby być na bieżąco z moimi wpisami i ofertami pracy w marce SAVICKI

——

Ich poste zwar viel, aber normalerweise nicht bei Empörungswellen.

Diesmal schon, weil ich merke, dass vor allem Frauen sich empören.

Was #CollienFernandes widerfahren ist, ist das extreme Ende eines Kontinuums, das in der Mitte der Gesellschaft beginnt: Kontrolle über Frauen durch Erniedrigung- digital, analog, privat, öffentlich.

Männer, die das schweigend lesen, sind leider kein Gegengewicht.

Volle Solidarität mit Collien und allen anderen Opfern von uns Männern!

PS: Wenn Du Dich als Mann mit eigenem Post unsicher fühlst: teile gern die Soli-Posts anderer oder diesen. Hauptsache Anteilnahme und ein Zeichen!

How we reinforce a positive culture in your company 📈

Epic: Someone built a Google Translate but for LinkedIn 👏😭

Whoever made this deserves a massive raise.

Absolutely brilliant.

BTW, on a serious note, in case you missed it, check out Andrej Karpathy’s Method To 10X Your Claude Skills 🧠: https://lnkd.in/dAg2yXVJ

This is so cool!
A group of high school students built a robot that can pick up balls and shoot them into a bin while moving without stopping, doing it with impressive speed and accuracy.

They call it Miss Daisy XXIV, created by a team from Wissahickon High School.

What looks simple in the clip is actually a full system working in sync. The robot collects balls on the move, stores them internally, then feeds them into a high-speed shooter that launches them precisely into the target.

It combines mechanical design, sensors, and software that has to make constant adjustments in real time while the robot is driving.

The barrier to building advanced robots is getting lower, while the capability is going up. That combination usually leads to one thing: rapid acceleration.

Pretty cool!

Follow me Endrit Restelica for more.

The most important habit you can have to succeed in your 9-5 and 5-9 🧠

"The role is labeled enterprise but it's really not enterprise sales."

Enterprise sellers typically earn >$300k OTE (usd). But many of them really aren't doing enterprise sales. Nothing inherently wrong with that to be clear, it's all just naming and titles.

Below you can see 120 sales orgs charted based on the average deal size vs. sales cycle length. (fyi, selling big deals in short periods of time is the holy grail!).

But are you really doing enterprise sales if you're not dealing with:

1/ >6 month sales cycles
2/ Multi-threaded buying committees
3/ Procurement and legal gatekeeping
4/ Custom pricing and deal structures
5/ Champion dependency
6/ Proof-of-concept requirements
7/ Heavy integration and security scrutiny
8/ Executive alignment
9/ Large ACV
10/ Spending cycles ensuring post-sale success.

And if you're interviewing for an enterprise sales role - many hiring managers will be looking for you to have a POV on these things and more, so...

We're tracking 100s of enterprise roles, have data on all of them like reviews, salaries, % of team hitting quota, etc.: https://lnkd.in/eZyEUFmZ

✌️

You are the sum of your daily actions.

If every day you get up and make good choices for your health, family, community, and career - you're going to see positive change.

If your everyday choices are poor you probably won't have the life you were expecting.

The daily habits that I try to do every day:
1. Breakfast with the family
2. Read Bible and workout (mostly push ups)
3. Focus on work and don't get distracted with other things
4. Spend time with Family in the afternoon

If the last week taught us anything, it’s this:

There are only a few true rockstars who can fill a stadium.

- Normal people → Taylor Swift 🎤
- AI people → Jensen Huang 🤖

Same energy. Different crowd.

#ai #humor

reminder that you’re not behind, it’s just moving too fast

Preparation is usually invisible.

Reading.
Thinking.
Learning.
Building relationships.

Then one day an opportunity appears.

And people call it luck.

You eventually start to realize, no job is safe.

The leaders who guard their spotlight the hardest
are the ones with the weakest light.

The manager who takes credit for every win.

The director who hoards information like currency.

The exec who sees talent as a threat.

They think protecting their position makes them powerful.

But it keeps their team small.

Insecure leaders don't understand that
helping others rise doesn't shrink their success.

It multiplies it.

When you lift people up:

→ They develop faster than you expected
→ They stay loyal when things get hard
→ They bring ideas they'd never share with someone protecting their ego

The leaders people remember aren't the ones 
who stood in the spotlight alone.

They're the ones who made room for others to stand in it.

Your light doesn't dim when you help someone else shine.

It spreads.

♻️ Repost to remind a leader to lift others
➕ Follow Dora Vanourek for more

More people are ditching ChatGPT for Claude than ever before.

But the way you prompt has to change as well.

What worked for ChatGPT won't necessarily get the same results in Claude.

This is what a bad, good, and great Claude prompt looks like:

❌ Bad Prompt

"Is this startup idea good?"

Why it's bad: 
Claude has no context. The output will be as vague as the input.

✅ Good Prompt

"Evaluate whether this startup idea is viable: [INSERT IDEA]"

Why it's better: 
It gives Claude something concrete to work with and a clear job to do.

✅ ✅ Great Prompt

Well, due to character count, you can find examples of these in the carousel. 

But as for why it's great:
Claude knows what you need, how to think through it, and exactly when to stop.

The "Great" prompt follows the structure my team uses across the board.

It includes:

1. Task context
2. Tone context
3. Background data, documents, and images
4. Detailed task description and rules
5. Examples
6. Conversation history
7. Immediate task description or request
8. Think step by step
9. Output formatting

Each prompt is a bridge between your ideas and the AI's capabilities. 

The more descriptive and structured you are, the better the results. 
And the more you can tap into the nearly limitless potential of Claude. 

There's a reason we built Searchable around Claude Code.
Compared to ChatGPT, it felt like we were building something extraordinary.

And after hitting $1.5 million ARR in 120 days... I’d say we made the right choice !

Searchable optimises for all the prompts listed in the carousel.  
And any others you can possibly think of...

If you want to ensure your business is visible in AI search, 
Start your 14-day free trial here: https://lnkd.in/epgXyFmi

How are you finding Claude compared to ChatGPT? Noticed a difference?
Share your thoughts below 👇

♻️ Repost to help your network improve their Claude prompts.
And follow Chris Donnelly for more guidance on leveraging AI.

Your future isn’t a mystery.

It’s already visible in your daily habits.

If you want to know who you’ll be in 5 years, don’t look at your goals. Look at this:

1. The books you read
↳ Choose books that really stretch you
↳ Highlight ruthlessly and turn ideas into action
↳ A great book changes behaviours

2. The foods you eat
↳ Your brain runs on what’s on your plate
↳ Eat 2–3 default “high-energy” meals per week
↳ Comfort food is just anxiety with better branding

3. The habits you build
↳ Tiny habits beat giant plans, you never start
↳ Attach new habits to old ones
↳ What you repeat in private = what you become in public

4. The friends you meet
↳ Your standards = the people you spend time with
↳ Ditch people who love gossip
↳ Spend time with people who talk about ideas

5. The workouts you do
↳ Training is for the 60-year-old you
↳ Treat workouts as investor meetings
↳ Strong body → calmer mind → better decisions

6. The money you invest
↳ Pay yourself first, then live on what’s left
↳ Buy assets & skills, not things to impress strangers
↳ Every $1 you invest buys you future freedom or regret

7. The sacrifices you make
↳ Saying “yes” to everything is the path to disaster
↳ Trade one hour of scrolling for one hour of building
↳ Short-term comfort is just an expensive addiction

Your future self is watching how you live this week.

Which one of these will you change today?

♻️ Repost it to inspire other leaders
And follow Andrea Petrone for more

(picture found on Pinterest)

Have you ever been told you are “too much”?

1) Too driven
2) Too honest
3) Too ambitious
4) Too expressive

Most of the time, the problem is not the person.

It is the environment.

Some spaces do not want growth.

They want comfort.

So they reward:

→ Blending in
→ Staying quiet
→ Keeping ideas small

When someone shows up with strong energy, they stand out immediately.

And standing out often gets labelled as the problem.

But strong ideas are not a weakness.

Clear communication is not a flaw.
Big energy does not need fixing.

In the right room, those traits move people forward.

The right people will not ask you to tone it down.

They will recognise the value in what you bring and make space for it.

Do you agree?

♻️ More people need to see this, share it with your network!

And follow Rob Dance for more LinkedIn content like this!

Prompting is the worst way to use Claude.

Here's what (actually) works:

Step 1: Download Claude (claude. com/ download)
Open Claude 'Cowork' tab. Select Opus 4.6 model.
Use this setup guide: https://lnkd.in/dVDent-3

Step 2: Open a Google Doc. Name it "about-me."
Write 4 things: who you are, how you write, examples of your best work, and your 'do nots.'

Step 3: OR copy-paste my interview prompt
From this guide: https://lnkd.in/eF56s4i8
It asks you 100 questions about how you think.

Step 4: Save it as .md and put it in one folder 
Download in .md format. Create a folder called "claude-context." Every .md file goes in there. 
Download my .md files here: how-to-ai.guide

Step 5: Upload the folder to Cowork
Click "Work with folder", upload the folder. Claude now reads every file before responding.

Step 6: Use this prompt to start any task:

"I want to [YOUR TASK] so that [WHAT SUCCESS LOOKS LIKE]. First, read the uploaded files before responding. DO NOT start executing yet. Instead, ask me clarifying questions so we can refine the approach together step by step. Only begin work once we've aligned."

Step 7: Use this prompt to redirect Claude:

"We are getting sidetracked. We need to start over without losing the context of this chat. First, ask me what went wrong and where we must go instead."

Step 8: Update your files as your taste evolves
Upload the old .md. Paste this prompt:

"Read this file. Update it with [what changed]."

After months of testing Claude, I built a single prompt library with every prompt I personally use.

To access it, complete these 4 steps:

1. Subscribe (for free) → how-to-ai.guide.
2. Open my welcome email.
3. Hit the automatic reply button inside.
4. Receive your prompt library + bonus video.

Contribute to the whole and you will likely be rewarded. Natural selection leads to better qualities being retained and passed along (e.g., in better genes, better abilities to nurture others, better products, etc.). The result is a constant cycle of improvement for the whole. #principleoftheday

We're all dopamine addicts.

The question is what kind?

Fast dopamine is easy. Scrolling, bingeing, it's junk food for your brain. The high hits hard. The crash hits even harder. And five minutes later, you're back looking for more.

Slow dopamine is different. It’s the buzz you feel after finishing a long book. It's the pride in cooking a meal from scratch. It asks more of you up front, but the reward actually lasts.

The problem is that most of us are so hooked on fast dopamine that slow dopamine barely registers anymore.

Silence feels uncomfortable. 
Boredom feels painful.

If something isn’t instantly stimulating, we bail.

So what do we do?

The solution isn’t to try and go cold turkey (trust me, I've tried).

It's to gradually shift where you’re sourcing your dopamine from.

This week, pick one fast dopamine habit and replace it with something from this list.

Just one swap.

Instead of scrolling before bed, read a chapter of a book.
Instead of popping in your headphones, try walking in silence.
Instead of ordering takeout, cook a meal at home.

Your brain will resist. It'll tell you this is boring, that you need your phone, that you should quit.

That's withdrawal.

Stay with it anyway.

Then next week, replace another one. And another.

You're not trying to be perfect. You're trying to retrain your system to recognize what satisfaction actually feels like.

The shift happens slowly. But it does happen.

When I was 27, a friend of mine made me feel very stupid about my $45,000 salary.

She didn't do it on purpose.

She was just genuinely surprised that I was "making so little" compared to what she was making at the same age.

It stung, to be honest.

Up until that moment, I hadn't really thought about whether my salary was good enough.

And I certainly didn't think it was a problem.

It's funny how ten minutes before this happened, I was doing fine.

I was content, even.

And 30 seconds after, I became fixated on how "far behind" I was.

How long would it take to close the gap?
Had I failed to reach my potential?
How did I get here??

The truth is that you feel behind because you only know how to read the current scoreboard.

And scoreboards are just snapshots of a specific moment.

They can tell you exactly where things stand right now, but they're terrible at predicting what happens next.

If you want to see how our journeys changed over the next 20 years, I share that in my new essay tomorrow morning.

You'll be surprised what "behing behind" can look like 20 years later.

Get it free here: https://buff.ly/A3EdMN4

This man is now worth $138 billion.

If Europe had at least one Steve Ballmer, we would easily overthrow Silicon Valley and become the tech & AI capital of the world.

Living legend.

🚀 Proud to share the latest issue of AGENTIC INTELLIGENCE - The world's first newsletter dedicated to AI Agents (and made by AI Agents!)

Receive this newsletter by email:https://lnkd.in/eqkb27tn

This Week's Agentic Breakthroughs:
1️⃣ Nvidia unloads at GTC with NemoClaw and more
2️⃣ Manus brings its AI agent to the desktop
⭐Forget the complex jargon about "automation-first thinking."
3️⃣ AI agent hacks McKinsey’s ‘Lilli’ chatbot in two hours
4️⃣ The Multi-Agent Trap | Towards Data Science
5️⃣ INTERPOL report warns of increasingly sophisticated global financial fraud threat
Which of these agent breakthroughs excites or concerns you the most? Share your thoughts below 👇

310,000+ pioneers have already subscribed to this newsletter. Why? Because very soon, we won't say, "There's an app for that." We'll say, "There's an agent for that."

#agenticAI #aiagents #ai #artificialintelligence

Today I turn 24

I think 18-year-old Aaron would be very, very proud.

Now I don’t want to do some AI “24 things I learned about sales” post.

Or some random dump, but I’ve been reflecting a ton this year.

On my journey, where I am and what my focus is moving forward.

So let me give you the mini rundown.

At 18 years old, I decided to skip university and bet on myself.

In 2020, I was signed to a major modelling agency in Manchester (ik right lol)

I wanted to follow in my brother's footsteps, who was crushing it.

He’d worked with Footasylum, Boohoo, ASOS, and had been flown around the world to Dubai, LA, Japan, Barcelona, and pretty much everywhere else.

As an 18-year-old from a council estate, that was THE DREAM.

Yeah I did a few music videos, but I never did well, partly because I was BROKE and couldn’t afford the shoots that I needed.

So I ended up in sales, and that same year I had built a team of 10 people and was a top rep in the entire country.

Now there are 1000 things between those 2 points but 1 thing is similar.

It was a commission-only role and again I had to bet on myself

And that’s been a theme of my life.

- I skipped university and bet on myself
- I quit a dead end job and bet on myself
- I quit a role I hated and bet on myself
- I didn’t invest in stocks heavily, I bought courses and bet on myself
- I quite an AE role, people thought I was crazy and bet on myself

Now at 24:

- I’ve been to 18 countries and can go wherever I want when I want
- I’ve built a business that is on track to hit multi 6 figure months very soon
- I’ve got a team of increidble people I geuinely love being around
- I’ve got a partner who has trusted in me more then I’ve ever trusted myself.

So if 18-year-old Aaron met me now, he’d be proud.

But the reason why I’ve been able to achieve anything is that I’ve bet on myself again and again.

If there’s something you want… go get it.

Bet on yourself, it’s the first step.

Life can move so fast.

So remember to take a minute and be proud of all you’ve done.

I’m sure the younger you would be proud too. ❤️

Anyways happy birthday to me lol

Ilekroć myślę, że oficjalny profil miasta Łódź (mojego miasta 🥹) niczym już mnie nie zaskoczy... wydarza się taki post, jak ten. 😅

I jeśli z miasta jednorożców 🦄 nie jesteście, możecie mieć wyzwanie, by poczuć, o co tutaj chodzi.

Spieszę więc z wyjaśnieniem.

Otóż profil Łodzi na Facebooku promuje akcję łódzkiego NFZ. 🩺 To chyba jest dość oczywiste.

Natomiast mniej oczywisty jest fakt, że robi to poprzez grafikę, przedstawiającą fontannę, która od lat budzi wiele... kontrowersji. 😬

Gdy tylko powstała, kojarzyła się mieszkańcom z żeńskimi narządami rozrodczymi. No i przyznajcie sami, że jak już Wam teraz tę myśl podsunęłam, to już tego nie odzobaczycie. Chyba, że sami na to wpadliście po sekundzie oglądania poniższej grafiki. Wtedy... wstydźcie się, jest piątkowy PORANEK! 😅

Tym samym profil miasta wykorzystał już istniejące skojarzenie i przekuł je w publikację w słusznej sprawie. 💡

Czy ta grafika przyciąga wzrok? No pewnie.

Czy budzi zainteresowanie? Liczba komentarzy, lajków i udostępnień (tu trochę parsknęłam) mówi sama za siebie. 😉

Czy dzięki tej publikacji ktoś zapisze się na badania? Bardzo na to liczę!

Podobnie jak na to, że więcej marek wyjmie komunikacyjny kij wiadomo skąd i stanie się bardziej ludzkimi. Może niekoniecznie jeżdżąc po bandzie jak profil Łodzi (bo też nie do każdej firmy czy instytucji to pasuje), ale jednak będąc ciut bliżej człowieka, dla którego się przecież tę komunikację tworzy. 😊

Dostałam zaproszenie na konferencję z adnotacją, że nadawczyni widziała mnie pięć lat wcześniej na innym wydarzeniu i cieszy się, że w końcu ma okazję mnie też zaprosić. I to mnie zmroziło.


Moja myśl: O MÓJ BOŻE, przecież ja pięć lat temu robiłam takie dzikie prezentacje, w ogóle nie umiałam robić slajdów, były żenujące i nie da się na te prezentacje patrzeć, ale wstyd.

Tymczasem Lewiński po wysłuchaniu tej historii: "ale czad, czyli zrobiłaś wtedy tak dobrą robotę, że pięć lat później wciąż ktoś to pamięta!".

Przeglądajcie się czasem w cudzych oczach. Tam bywa milej, niż we własnej głowie.

Kolejny tydzień życia za nami.
I jaki to był tydzień?

#PracaPrzyPiątku

Skrin pochodzi od Krzysztof Kotkowicz, pod postem Martyna Tarnawska - jest tak dobry, że zasłużył na kolejną rundę po Linkedinie.

over the last 10 years, l've often felt lonely, isolated, and at times stressed. One of my antidotes has been my dog…Pablo

Of course Dogs are known for their unconditional love, but as a founder Pablo has made me remember that life can be simple - he's helped me realise that most of my worries don't actually matter.

After a difficult day, watching him play with a bottle cap or get excited by a stick has been an interesting reminder that we can find happiness and joy in simple things and that much of our struggle is a consequences of our own overcomplicated perspective, overthinking and self-focus.

Pablo is now entering the final season of his life and has some walking issues. I know it might sound strange - especially to someone who's never had a dog - but I wish he could understand how much he's done for me, how deeply he's helped me at times, and how much I'll miss him when he's gone.

Neuer Job.

Gleiche Mission.

Ab heute: Beiratsmitglied bei #SheTransformsIT.

Danke, Pauline Meimberg, für die Gelegenheit und schön, dass Ihr - Bitkom - die Initiative weiterhin unterstützt!

Für mich neues Feld, doch das Thema bleibt: Potenziale sichtbar machen.

Yes, she can. For sure!

How to organise a dream conference:
A 10-step masterclass by Joe and James
 
Step 1: Recruit an army of event volunteers comprised exclusively of the world’s warmest people (how did you do that?!)

Step 2: Arrange beautiful weather so the walk to the venue feels like you’re in a scene from an American romcom based in London

Step 3: Pick a stunning venue practically next door to Westminster Abbey and Big Ben, and decorate it beautifully. Oh - and play bird song, it’s really calming. 

Step 4: Somehow manage to treat every single one of the 400 attendees like they’re the guest of honour everyone has been waiting for.

Step 5: Brilliant food. Top-notch hot lunch. And snacks. Snacks everywhere. Have volunteers that actively encourage greedy people (OK, me) shoving handfuls into their bags.

Step 6: Have a screen (and an app option) with live, fast and accurate subtitles for those who struggle with hearing what's happening on stage

Step 7: Allocate a comfy quiet room. Conferences can be so overwhelming and having 5 minutes quiet can make all the difference. 

Step 8: Curate the best and brightest line-up of speakers ever: 

• The hilarious Dave Harland
• My new hero Mimi Turner
• Completely captivating Maddy Allen
• Camp Nou architect Phill Agnew
• Absolute badass chima mmeje🏳️‍🌈
• Squiggly and inspiring Helen Tupper
• Cool and insightful Naomi Walkland
• Creative genius Lee Bofkin DPhil FRSA
• Fascinating uncertainty expert Sam Conniff
  
Step 9: Oh, and have professional speaker coaches on hand to offer breathwork and stage presence support throughout the day (Maddy, you’re amazing). 

Step 10: Last but not least, make sure your organisation is led by two of the most sincerely lovely people you’ll ever meet in your life: Joe and James. 

A huge thank you to all the epic The Marketing Meetup, showing you CAN have a large, successful business (they’ve got 211k members now!) without losing kindness and humanity. 

You’re genuinely changing the world and I’m so grateful to have any part in it.  ❤️

PS Post about my own talk and AI letting me down coming next week when I've stopped cringing 🤣

Weniger Meetings = mehr Gehalt für Mitarbeiter.

Die meisten Meetings bringen keinen echten Mehrwert. Im Gegenteil, sie kosten Fokus, Zeit und Energie. Und vor allem kosten sie eines: Geld.
 
Das zeigt eine Studie von Steven Rogelberg: 25.000 Dollar pro Jahr - pro Mitarbeiter - gehen durch unnötige Meetings verloren. Bei Führungskräften sogar 42.000 Dollar.
 
Hochgerechnet auf ein Unternehmen mit 5.000 Mitarbeitern sind das über 100 Millionen Dollar.
 
Das muss man sich mal vorstellen. Warum schaffen wir unnötige Meetings nicht ab und stecken dieses Geld in die Gehälter?
 
Rechnen wir das mal durch:
Aktuelles Gehalt: 60.000 € + „Meeting-Verschwendung“: ~23.000 €
= 83.000 € mögliches Gehalt
 
Selbst wenn man nur die Hälfte umverteilt, sind das 11.500 € mehr Lohn pro Mitarbeiter pro Jahr.
 
Jedes Meeting, das keinen echten Mehrwert bringt, muss konsequent gestrichen werden. Bezahlt wird Leistung, nicht das Absitzen von Terminen.
 
Alles andere ist nicht nur ineffizient und kostspielig. Es ist respektlos gegenüber der Leistung und der Zeit der Mitarbeiter.
 
Weniger Meetings ➡️ mehr Fokus ➡️ bessere Ergebnisse ➡️ mehr Wertschöpfung.
 
Und genau so entstehen höhere Gehälter.
 
Wie viele Meetings in eurem Kalender haben wirklich Mehrwert?

#meetings #kosten #gehalt #mitarbeiterführung

Stop overthinking...it's driving you crazy

We all know them.

And over time, you learn to spot them quickly.

The ones who only show up when they need something.

Don’t be that person.

Make it a habit to reach out to people when you don’t need anything 💯

Agree?

Die zwischenmenschlichen Kosten für Widerspruch sind höher als die für schlechte Ergebnisse.

Wer sich auflehnt und Dinge kritisch hinterfragt, wird im Business oft sozial beschämt, abgewertet oder als unbequem gelabelt.

Immerhin wollen wir alle mit unseren Aufgaben fertig werden und stehen unter Druck. Wer kann da schon Störungen gebrauchen.

Wir nehmen lieber hin, dass wir:
👉🏼bis zu 41 % unserer Arbeitszeit mit sinnlosen Tätigkeiten verbringen (Deloitte Human Capital Trends, 2025)
👉🏼häufig sichtbaren Aufgaben den Vortritt geben, egal ob sie sinnvoll sind (Visier Studie, 2023)
👉🏼in der Führung oft Fehler gar nicht oder nur teilweise zugeben (EY Fehlerkultur Report, 2023)

Aber wir können uns ein „Hauptsache fertig werden – egal was dabei rauskommt“, nicht mehr leisten.

Deshalb müssen wir Wege finden, wie wir einander widersprechen können und um die besten Lösungen streiten, ohne dass es uns unseren sozialen Status kostet.

Wie das möglich wird, beschreibe ich im neuen Buch von Emma-Isadora Hagen.

Außerdem finden sich darin 99 weitere richtig starke Impulse, die zeigen, wie wir Arbeit heute gestalten müssen, um gesund und erfolgreich sein zu können.

👉🏼Es erscheint am 24.03. und ist schon vorbestellbar.

China doesn't need to win a war. It just needs America to stay busy fighting one.

In our recent episode, Khan Sir talked about how China is benefiting from the current global conflict.

America is currently stretched thin. Iran, domestic fractures, economic pressure on multiple fronts.

And while that's happening, China is deepening its grip on neighbouring countries.

Taiwan, home to the world's most advanced semiconductor manufacturing, is the real target. The moment America is overextended, the window opens.

And there's also the part that America's foreign policy isn't shaped by presidents alone. It's shaped by who funds the elections.

Khan Sir's take was that Israel is the largest funder of US elections. Silicon Valley to Wall Street, the lobbying network runs deep. Decisions that look like presidential choices are often systemic outcomes.

So where does that leave India?
"We don't fear anyone. But we don't make anyone fear us either," Khan Sir said.

India has relationships with everyone. Russia. America. Israel. The Gulf. That's valuable.

But relationships without leverage just mean you're invited to every table without controlling the decisions.

The question for the next decade isn't who India's friends are. It's whether India can shape outcomes.

Full episode out now. Link in comments.

This episode is delivered by Porter.

#rajshamani #figuringout

Eid Mubarak an alle, die heute feiern.

Und ein ehrliches Danke an alle, die nicht selbst feiern und trotzdem gratulieren.

Es sind oft die kleinen Gesten, die zeigen, wie wir miteinander leben. Eine Nachricht, ein kurzer Wunsch oder ein Moment des Mitfreuens sind stärkende und verbindende Momente.

Gerade in einer Zeit, in der ungefilterte Boshaftigkeit im Netz so präsent ist, bekommen diese persönlichen Begegnungen ein anderes Gewicht.

Sie erinnern daran, dass Gesellschaft nicht aus Kommentaren besteht, sondern aus Menschen.

Und vielleicht kennen viele dieses kurze Zögern: Darf ich das überhaupt sagen?

Oft sind es genau diese kleinen Zeichen von Zugewandtheit, die mehr verbinden, als man denkt.

Most people will never master AI (properly).

But these 7 skills make it embarrassingly simple:

Skill 1. Stay updated with AI News
✦ Pick 2-3 creators who teach AI step-by-step.
✦ One newsletter. Once a week. That's it.
✦ Every article you read → try one thing. Now.
✦ Start with my free guide here: how-to-ai.guide.

Skill 2. Pick one AI tool and master it
✦ Pick one. Delete the rest from your bookmarks.
✦ Use it for 30 days. Only that tool. Go deep.
✦ Learn its Projects, memory, search & file uploads.
✦ Copy my project guide: https://lnkd.in/exfaYnBQ

Skill 3. Set up your AI before you prompt
✦ Create a folder: "AI Files."
✦ First file: who you are, your tone, your audience.
✦ Upload files → define task → define success.
✦ My full setup guide: https://lnkd.in/exfaYnBQ

Skill 4. Teach AI what you know
✦ Prompt: "Ask me questions about my expertise."
✦ Let it extract your rules, your no's, your audience.
✦ Export into one .md file. Reuse it for months.
✦ Copy my .md file prompt: https://lnkd.in/eF56s4i8

Skill 5. Talk to AI like a colleague
✦ Start with "Don't start yet. Ask me questions."
✦ Read v1. Name all the things that are wrong.
✦ Push harder: "Argue against this."

Skill 6. Ship before it's perfect
✦ Build the rough draft with AI in 20 minutes.
✦ Show it. Let people react to something real.
✦ Sell what it becomes once they invest.

Skill 7. Lead AI. Don't follow it
✦ Split every task: what does AI do? What do I do?
✦ Give AI the 80%. Keep the 20%.
✦ If you can't spot the mistake, don't delegate it.

The skills that matter in 2030:

The ones AI makes 10x more powerful.
Or can't touch at all. The middle gets automated.

To learn more, read my free newsletter:
https://lnkd.in/ebYZFYx3

♻️ Repost if your network needs this wake-up call.

A new beginning is not just about starting again.

It is about becoming someone who will not repeat the same mistakes.

A New Name. The Same Trusted Team.

After many years operating as HLB Mann Judd Insolvency WA, I’m excited to share that our practice has transitioned to a new name:

Equinox Restructuring & Insolvency

This marks the beginning of a new chapter for our team as we move forward under our own independent identity.

Our time as part of the HLB Mann Judd Australasian Association has been an important part of our journey. The network and professional relationships built during that time helped shape the practice we are today, and we are grateful for the collaboration and support we experienced along the way.

While the name is changing, the fundamentals of our practice remain exactly the same:

✔ The same people
✔ The same expertise
✔ The same capacity
✔ The same practical and respectful approach

Our team will continue to support accountants, lawyers, financiers and business owners across Western Australia with restructuring, turnaround and insolvency matters.

The name Equinox represents balance and transition — a point where one season shifts into the next. It reflects the work we do every day: helping businesses and stakeholders navigate challenging periods and move forward with clarity and confidence.

If you'd like to learn more about our new brand, please visit our new website:

www.equinoxri.com.au

I’m incredibly proud of the team and excited about the next phase of our journey.

Equinox Restructuring & Insolvency
Formerly HLB Mann Judd Insolvency WA

Thank you to everyone who has supported our practice over the years. We look forward to continuing to work with you.

#equinoxri #restructuring #insolvency #businessrecovery #turnaround #accounting #law #perthbusiness

One of the most underrated career skills:

Never stop learning.

I see many people stall because of the opposite.
They do stop learning, and they get a little too comfy.

(Credit: Thomas Pearce)

The reality is...
If you aren't willing to get uncomfortable, you aren't going to grow.

That includes learning in your 5-9 and weekends.

Which can sound like a LOT to some people.
But to be honest, if it does, you aren't a high performer.

Don't expect great results without great effort.

1. Set delusional goals.
↳ In order to learn, you need to have goals first.
↳ Neuroscience says the more delusional the better!

2. Capture ideas in your Notes app.
↳ This has honestly changed everything for me.
↳ Take notes as they come to you, and expand on it later.

3. Take time for tiny reading sessions.
↳ In a spare 5-10 minutes, don't open Insta.
↳ Read a fic or non-fic book instead!

4. Mimic the very best.
↳ Look at someone who has achieved what you want to.
↳ Follow their blueprint and avoid their mistakes.

5. Utilise Repeat Exposure.
↳ You don't remember stuff you learned once.
↳ Repeat the learning and your brain will remember it.

6. Sacrifice your free time.
↳ I'm not saying don't have fun, not at all.
↳ But spending too much time on that limits growth.

The compound of all of these is incredible.

Lock the F in for a year every single day...
The person you'll look back at will be unrecognisable.

What an honor and a privilege to receive the Legacy award from Black Enterprise! I was thrilled to beyond words. Thank you to Butch, Alisa, Caroline, Genevieve and the rest of the leadership team! I am also so honored to be in the class with Angela Bassett, Rosalind Brewer, Shellye Archambeau, and Bennie Wiley! Icon, icon, icon, icon! God is good all the time and all of the time God is good! #BEWPS, #CarlasPearls
#ExpectToWin
#PressureMakesPearls

Kolejna bariera pęka w UK 🇬🇧

Najlepsze od 17 sekundy ...🇵🇱

InPost UK ❤️

A few months ago, I interviewed a candidate who bombed the interview.

It would have been easy to reject her, but I went with my gut and gave her a chance.

And she’s killing it. ❤️

Some of the strongest hires:

💡Think deeply before they speak
💡Aren’t great at “selling themselves”
💡Get nervous when everything is on the line
💡Ask thoughtful questions instead of rushing answers

The interview didn’t reflect the employee she became.
It reflected an amazing human being under pressure.

A reminder to hiring teams:

Don’t be too quick to dismiss someone who doesn’t interview flawlessly.
Sometimes the best hires don’t “kill” the interview.

Instead, they shine when leaders look beyond the moment and hire for potential, not polish.

🍀 Please share to promote giving people a chance.

Wild: Goldman Sachs dropped the most uncomfortable AI stat yet - $450 billion was spent on artificial intelligence last year, but it contributed basically zero to economic growth 😳

All of 2025 was supposed to be the “AI boost” year.
Instead, the data says… nothing happened.

But not because AI is fake.

Because the money didn’t go where people think.

↳ $130B+ went to Nvidia
(we built the shovels, not the gold)

↳ Massive buybacks
(layoffs → higher margins → stock up)

↳ Imported hardware
(GPUs built in Asia = cancels out in GDP)

↳ Early productivity gains
(real, but trapped inside teams - not the economy)

Meanwhile:

→ Meta cut 21,000
→ Atlassian cut 1,600
→ Block cut ~40% of staff
→ Amazon keeps automating warehouses

The jobs moved fast. The growth didn’t.

We’re now in the “burn first, benefit later” phase of AI.

Companies restructured ahead of the payoff.
Markets priced in the upside.
But the actual economy hasn’t caught up.

Historically, this lag is normal.

Electricity, the internet - same pattern.
3-5+ years before productivity shows up in GDP.

But the difference this time is that the layoffs came first.

So now we’re sitting on a strange moment:

Record AI spend.
Record infra buildout.
Very real job displacement.

And basically no GDP growth…

This is either the setup for a boom or the cleanest bubble we’ve seen in years.

Żeby wytłumaczyć, jak powstało to wydarzenie, muszę zacząć od wyjaśnienia dziwnego słowa:

Serendipity

to proces „inteligentnego szczęścia”. Kluczowe kroki to: zauważenie nieoczekiwanego wyzwalacza, reinterpretacja danych i możliwości oraz podjęcie działania dzięki przenikliwości i wytrwałości.

Tak się ostatnio przydarzyło, że pozbawiono mnie możliwości wygłoszenia prezentacji, w którą włożylem dużo pracy. Jednak publiczność, która (w przeciwieństwie do mnie) dotarła na tę konferencję, bardzo chciała mój materiał zobaczyć. Zadałem sobie pytanie: po co marudzić na rzeczywistość, na którą nie mamy wpływu? Lepiej tworzyć własną.

Tak oto powstała pierwsza konferencja z serii „Kozieł i Przyjaciele”, na którą zamierzam zapraszać moich mądrych znajomych oraz wypytywać ich o ciekawe rzeczy.

Na pierwszy ogień bierzemy storytelling AI, o którym pogadam z naprawdę świetną ekipą.

Polecam dołączyć, jeżeli:
- występujesz publicznie,
- budujesz markę osobistą albo firmę,
- prowadzisz social media, podcast lub prezentacje,
- chcesz używać AI w sposób praktyczny, a nie tylko efektowny,
- masz dość generycznych treści i szukasz sposobu, by lepiej mówić, pisać i sprzedawać.

Konferencja odbędzie się 1 kwietnia o 16:45, co brzmi jak żart, ale zapewniam, że to poważna sprawa.
 
Jeżeli ktoś chce wpaść osobiście, to zapraszam do siedziby Just Join IT w Warszawie, przy Marszałkowskiej 107 (wymagana rejestracja, miejsca ograniczone).
Jeżeli wolisz obejrzeć online, to wystarczy się zarejestrować na: kozieliprzyjaciele.pl

If you’re an AI engineer working on RAG, or building advanced retrieval-augmented systems, you need to know about RAFT: Retrieval-Augmented Fine-Tuning.

Let’s break it down 👇

→ Closed-Book Models (SFT Only)
The model learns everything at train time, and answers based purely on its internal weights. Fast, but brittle – hallucinations spike when the model faces unfamiliar queries.

→ Open-Book Models (Standard RAG)
At inference time, the model retrieves top-k documents and answers using them as context. But the model has never seen these docs during training – so it treats relevant and irrelevant documents the same way, often leading to noisy outputs.

→ RAFT: Retrieval + Fine-Tuning Combined
RAFT, proposed by UC Berkley, merges RAG and fine-tuning. During training, the model is explicitly taught how to use retrieved documents – rewarding it for grounding answers in the right document and ignoring distractors.

Here’s how RAFT works:
→ Use a query
→ Pair it with a golden doc (the correct reference)
→ Add sampled negative docs (distractors)
→ Train the model to generate an answer that quotes only from the golden doc

This makes the model retrieval-aware during generation – it learns to differentiate between helpful and irrelevant documents.

Why RAFT matters 🤔
→ Reduces hallucinations by grounding answers in relevant context
→ Boosts accuracy in domain-specific applications like legal, medical, scientific QA
→ Works with smaller open-weight models like LLaMA 2 and Mistral 7B
→ Outperforms vanilla RAG on benchmarks like HotpotQA and PubMedQA

How to train with RAFT 🛠️
→ Build training triples: (query, golden doc, distractor docs)
→ Use your existing retrieval setup and corpus
→ Fine-tune using LoRA or full SFT with these inputs
→ At inference, continue to use top-k retrieval – the model will now handle noise better

When to use RAFT ⁉️
→ When your application requires faithfulness and traceability (e.g., legal, healthcare)
→ When your retrieval corpus includes overlapping or ambiguous docs
→ When you want smaller models to reason better with external documents

RAFT doesn’t replace retrieval – it enhances it by teaching the model how to reason over retrieved content.

Instead of hoping your model figures it out at runtime, RAFT prepares it during training.

If you’re working on GenAI systems or retrieval pipelines, this is one method you can’t afford to ignore.

Arvind and I are doing a free RAG lightning session on 4th April. If you want to learn more about RAG, do join us: https://lnkd.in/gHFmmfR2

After 30 years studying human behavior, here’s the quiet truth:

Most people don’t think poorly because they lack intelligence. They think poorly because their environment keeps pulling their attention apart.

Fix the environment, and clearer thinking follows.

agree?