Mohammad Shaker's Blog, page 2

KnowING What I Can’t Control.

When should I adopt a new technology? When should I take risks? When should I play it safe?

It all ties to risk. Ten years after reading and rereading Nassim Nicholas Taleb’s work, I think I understand a bit more.

I must know what falls within my circle of control. More importantly, I must know what I cannot control.

This view of looking at everything from the prism of risk is echoed by the thinkers I admire—Hayek, Taleb, and Seneca. In economics, in daily life – and in what I try to do in tech.

12 years ago I learned the first step:
“Focus on what’s under my control and leave what is not.”

10 years ago I learned to take it one step further:
“Focus on what’s under my control and render what’s not ineffective” (Antifragile, Taleb)

This is the essence of Taleb’s barbell strategy. In investing, keep 90 % of capital in ultra-secure assets and put the remaining 10 % in highly speculative bets with unlimited upside (and downside). However that 10 % fares, you won’t be ruined; your tranquility remains intact.

In tech and startups, the analogy is to design systems that are ultra-stable—using technologies the team knows and trusts—while reserving 10 % for bleeding-edge experiments. Juniors often disagree; seniors rarely do. Juniors get excited about every shiny tool without proof of its durability (a la the Lindy effect).

The barbell strategy doesn’t encourage timidity; it limits the cost of being wrong. I want to survive—and to do so wisely.

I’m not chasing hyper-growth;

I’m chasing sustainable growth,

in both

business

and

life.

Know What I Can Not Control.

When to adopt a new technology? When to be risky? When to stay safe?

It all ties to risk. 10 years after reading, and re-reading, Nassim Taleb (NNT) works, now I think I understand a bit more.

A Framework for Control

I should know what I can control, what falls in my circle of control. More importantly, I should know what I can not control.

This is echoed by most of the people I love and idolize: Hayek, NNT and Seneca. One of key ideas in stoicism is to deflect unnecessary distraction. To focus on what matters and leave what doesn’t.

What I didn’t know 12 years ago is that I should take this a step further:

To focus on what’s under my control, and leave what is not.

And what I didn’t know 10 years ago is that I should take this another step further:

To focus on what’s under my control, and render what’s not ineffective (Antifragile, NNT)

Same idea of Barbel Strategy by NNT and risk. An example in investing by NNT is to keep 90% of one’s investment in ultra secure funds and leave the rest of the 10% in highly speculative bets with unbounded upside (and downside). Either way the 10% goes, you’ll not be ruined. You’ll stay in your tranquility.

Another example in tech is use of technology and startups. I should always try and design systems that are ultra stable, use technologies, we as a team, know and confident about and leave 10% to try ultra new, ultra edge tech – experiment. Many would disagree but I think they would be the newbies. None of the people older (and more wiser) than me agree with me. You can see it in juniors that they get extra excited about every new technology out there without the slightest proof of its validity (Lindy Effect.)

This Barbel strategy doesn’t encourage being safe. It simply limits the risk of being wrong. I try to survive. I try to be wise doing that. I’m not chasing ultra growth. I’m chasing sustainable growth – whether this is in business or in personal matters.

Salam, peace.

“I’m not a Hypocrite.”

I should only ask people to do things I will do myself. If I’m asking someone to do something that I won’t do myself, then I am a hypocrite. Plain and simple.

The Golden Rule by Hammurabi says: “Do unto others as you would have them do unto you.”

The Egyptian, the Indian, the Greek, the Romans, and the Persian; they all share a version of this Golden Rule.

The inverse, “If you don’t want “X” done to you, don’t do “X” to someone else”, is the Silver Rule*: “Do not treat others the way you would not like them to treat you.”

The Silver Rule translates well in business: Never ask someone to do something you won’t do yourself. Never do to others what you don’t want them to do to you. Do to others what you want them to do to you.

This served me well; especially working in small startups early in my career. I should only ask people to do things I will do myself. If I’m asking someone to do something that I won’t do myself, then I am a hypocrite. Plain and simple.

When I started my first two startups, we were bootstrapped and we were resource scarce. The team was a small team of fresh grad across the board: engineers or (a) marketer. That’s what we could afford. They were juniors who had great potential. None of the engineers has worked with proper cloud provider before (like AWS) for instance (10 years ago)

What made this team a killer is the attitude. We launched our first full webapp and a full working backend in 3 months. We all showed our skin in the game. We were all hungry to learn and none was a BSer.

Skin in the Game

I’m still extra proud of working with such a team. I was completely engrossed with the work with everyone in the company. I wasn’t a CEO or a CTO or a CFO or a CMO. I was all of these. I worked with everyone side by side. I had skin in the game and I showed it.

I did everything. I programmed the first backend service. I designed our app and the website UI. I fixed bugs myself. I created, designed and posted IG posts just like any marketer. I answered customers emails and I reached out to influencers. I (cold) emailed school heads just like any sales person.

I did everything I would ask a designer, an engineer, a marketer, or a sales rep to do. I showed them that I’m there for them. And they showed me their best.

It wasn’t all rosy. I made every managerial mistake in the book. I got angry, I put someone down on public, I was impatient when I should’ve and patient when I shouldn’t have.

Though, the good thing that I did was that I told them I’m wrong when I’m wrong. I had regular 1:1s and reverse 1:1 weekly with all of them. I kept everything open and everyone can speak out his mind.

I simply learned management on the spot. By practice. Bottom up. Not leading from the front nor from behind. But side by side.

Thanks

To teach is to do. And no one would’ve trusted me in hard times if I didn’t do that.

And hard times we had – just like any startup: limited runway, wrong targeting and a pivot within 1 year.

In a 3-year period we lost none on the team. And I believe this is because we trusted each other in bad times the same as in the good times.

People need to trust us for them to work with us. If we show them that we understand their pains, they would understand that we’re real colleagues – authentic and trustworthy.

These two traits are rare,

extremely rare, nowadays.

No one likes to work with someone they don’t trust. No one. If you’re a manager who can’t code, none of the engineers can trust you. If you’re a head designer who can’t design literally everything (as Massimo Vignelli would say: “To Design is to Design Everything”), none of the designers will trust you. If you’re a founder who’s undecided, no employee will trust you.

Skin in the game that is. And in the game; skin, sweat, tears and blood will be spilled. And the people we work with will reward us for seeing us sweat with them.

Salam, peace.

* Skin in the Game, Nassim Taleb.

“Requirement” is an empty word.

“Requirement” is an empty word. It’s overused and it does mean different things for different people.

Take a problem, P, with a requirement, R in a software engineering context.

If I say something R is a requirement to solve P, everyone would think that:

I know what I am doing.I know what I want.

I simply narrowed the solution space because I have forced everybody to take the narrow perspective of the requirement R of the overall problem P.

It’s a “requirement” after all. I do “require” it. Or the company does “require” it. Or the customer does “require” it. It’s nearly a non-negotiable.

The more senior you are, the more people will see a “requirement” as “commandment.” You have your holy text with you that no body should or can question.

Avoid it at all cost unless you’re 100% sure that something is truly required. Keep the solution space open with User Stories and Narratives instead. We use this in all our [Product/Technical] tickets on Asana in SpatialX.

Salam, Peace.

The Current Expert Problem with ChatGPT

I was talking with my friends, Mehdi Zonji about this. And I thought I should share.

ChatGPT’s ability to generate responses across a vast range of topics is both a strength and a limitation. It’s not a sniper (answers specifically), it’s a shotgun (answers generically.) And that’s a problem.

On the technical side, ChatGPT does a really good job of a “Junior Engineer.” As the time of writing in Q1 2025, ChatGPT still misses the 2nd or 3rd order effect of its proposed technical solutions.

Take ChatGPT’s confidence in incorrect answers. Right now, ChatGPT is like a junior who never says “I don’t know.” It will give you an answer with confidence, whether or not it should and whether it’s indeed the right answer or not.

This is extremely dangerous for new tech engineers coming into the field where they copy/paste solutions (and mostly code) without understanding what’s inside. For research and experimental work, this is fine. For things to go to prod, this is not.

ChatGPT still is incredibly useful—as long as you treat it like a junior team member who needs review, context, and guardrails. And you, the senior engineer, still need to nudge it to use a specific design patterns, or use of OOP or correct it that a Lambda function has memory and execution limit and that the solution it proposed is not valid.

In Product Management, ChatGPT can help with surface-level tasks: generating user stories, writing PRDs, summarising customer interviews. That’s helpful, and in many ways mirrors a Junior PM—someone who’s learning the ropes and can produce artifacts quickly if given direction.

But it can’t replace the judgment of a Senior PM who’s making trade-offs between business value and tech complexity, who knows when to ship a rough cut vs. when to push for polish, and who aligns stakeholders without just regurgitating frameworks.

For example, if you ask ChatGPT how to prioritize features, it might give you RICE or MoSCoW. But it won’t challenge the underlying assumptions, like: Are we even building the right product? What’s the risk of not shipping this? It can’t feel the tension between short-term growth and long-term strategy. That requires context, lived experience, political awareness, and an instinct for timing that machines currently don’t have.

The danger is when people start treating ChatGPT like a senior.

When it comes to deep expertise, a human, a senior, is still needed.

This will change. And I hope it does.

Weekly: 3 biggest problems(3BP)

Here’s a digest of my week.

Something I want to keep doingGym 5-6 times/week.Focus on 1 thing in each day. Not thinking about anything else. I do plan the next day at the end of each day.Sitting and Writing.Something I’m learning.Technical: How Figma does sync its events between the frontend and the backend. We’re not using something similar in SpatialX.Technical again. VectorDB.Something I’m remembering from the past

My father, a lot.

My current exercise regimen120 steps (on a stepper) in 20 minutes.40-60 min weights training.No runs sadly this Ramadan.My current day routine regimenUp.Read for an hour.Work till 6.Gym.Iftar and sit with the family.Reading/working for 2hrs.Sleep.What’s working for you that you’d be crazy to change?Focusing on 1 thing at a time.Asking myself what are the 3 biggest problems (I call it 3BP) that I want to solve: whether this is daily or weekly.Planning for the week ahead. Contrary to what most people think, the idea is not to fill the calendar, but to clear the calendar when I plan.Solve the biggest problem first.What’s not working for you, and you’d be crazy not to change?Not reading enoughNot writing enoughNot sharing technical work and the great things we’re doing technically at SpatialX.What makes me waking up enthusiastic lately? What can I do this week or month around this?Our work at SpatialX really fires me up. Seeing it come together really makes one proud. Thanks to every single person in our small team.

That’s it for the week.

Weekly: 3 biggest problems (3BP)

Here’s a digest of my week.

Something I want to keep doingGym 5-6 times/week.Focus on 1 thing in each day. Not thinking about anything else. I do plan the next day at the end of each day.Sitting and Writing.Something I’m learning.Technical: How Figma does sync its events between the frontend and the backend. We’re not using something similar in SpatialX.Technical again. VectorDB.Something I’m remembering from the past

My father, a lot.

My current exercise regimen120 steps (on a stepper) in 20 minutes.40-60 min weights training.No runs sadly this Ramadan.My current day routine regimenUp.Read for an hour.Work till 6.Gym.Iftar and sit with the family.Reading/working for 2hrs.Sleep.What’s working for you that you’d be crazy to change?Focusing on 1 thing at a time.Asking myself what are the 3 biggest problems (I call it 3BP) that I want to solve: whether this is daily or weekly.Planning for the week ahead. Contrary to what most people think, the idea is not to fill the calendar, but to clear the calendar when I plan.Solve the biggest problem first.What’s not working for you, and you’d be crazy not to change?Not reading enoughNot writing enoughNot sharing technical work and the great things we’re doing technically at SpatialX.What makes me waking up enthusiastic lately? What can I do this week or month around this?Our work at SpatialX really fires me up. Seeing it come together really makes one proud. Thanks to every single person in our small team.

That’s it for the week.

On Trust.

Writing this reminds me of when I started working on my own two startups seven years ago. I struggled to delegate work. I wanted everything done to my standards—my quality, my ethics. In my mind, I was the best at what I did—“no one does it like I do.”

I think many capable engineers go through this. We start with zero trust in others. Slowly, we learn to trust people to do their jobs as well as we do—sometimes even better.

It’s a mindset shift—realizing that others bring different approaches and that we can actually learn from them. But the truth is, exceptional engineers are rare. The more we demand excellence from ourselves and others, the harder it becomes to trust that anyone else will meet that standard.

Trust isn’t given—it’s earned. No one deserves a free pass because of their title or past experience.

Trust is earned,

every time,

anew,

in every new venture we create or join.

Salam, peace.

On Trust

Writing this reminds me of when I started working on my own two startups seven years ago. I struggled to delegate work. I wanted everything done to my standards—my quality, my ethics. In my mind, I was the best at what I did—“no one does it like I do.”

I think many capable engineers go through this. We start with zero trust in others. Slowly, we learn to trust people to do their jobs as well as we do—sometimes even better.

It’s a mindset shift—realizing that others bring different approaches and that we can actually learn from them. But the truth is, exceptional engineers are rare. The more we demand excellence from ourselves and others, the harder it becomes to trust that anyone else will meet that standard.

Trust isn’t given—it’s earned. No one deserves a free pass because of their title or past experience.

Trust is earned,

every time,

from scratch,

in every new venture we create or join.

Salam, peace.

On Self Worth.

Waiting for outside approval is one of the worst thing anyone can tie himself into. People who wants/strive/look-for outside validation, approval, appraisals are weak.

If we do not have the self-worth from the inside, no outside validation can work. If we simply follow the herd in “Fake it till you make it”, we are simply cheating ourselves. “Don’t fool yourself and you’re the easiest man to fool.”

We are the ones who decides our self worth. We should be the ones who self-question, self-criticize (Popper), self-validate, and self-improve.

Most people tie in the idea of effort to the idea of self worth and validation. In the [imaginary] realm of schools and universities, if we study well, we’ll pass. We’ll succeed. We’ll praised. In the real world, and in business, if we work hard, if we spend a lot, we may not succeed. But that should not undermine our self worth.

Success or failure is not what determine our self worth. We do. Trying, doing, acting, tinkering is honorable.

Salam, peace.