Last Thursday I was debugging a race condition in a WebSocket handler when my phone buzzed. A message from a friend: "Did you do the trigonometry homework?"
I looked at the clock. It was 11 PM. I had not done the trigonometry homework. I had also not reviewed the three pull requests sitting in my inbox, not replied to an email about a commercial partnership, and not studied for the AYT chemistry mock scheduled for Friday morning.
This is what context switching looks like in practice. Not the clean, productivity-blog version where you time-block your calendar and meditate between tasks. The actual version, where everything is due simultaneously and you are one person.
For context: the YKS is Turkey's national university entrance exam. A single test, taken on a single day, that determines which university you attend. Millions of students sit it every year. Preparation typically devours the last two years of secondary school. Students attend private tutoring centres. They study past midnight. They give up sports, friends, hobbies. The entire social structure of Turkish adolescence bends around this exam.
I am doing that. I am also chairing a multinational holding company with four operating divisions.
The obvious question is: how?
The honest answer is: imperfectly.
There is a popular theory in productivity writing that context switching is the enemy. That you should batch similar tasks together, maintain flow states, protect deep work. The theory is correct, in the same way that a physics textbook is correct when it describes a frictionless surface. True in principle, useless in the conditions where you actually live.
My days do not permit batching. A school day runs from eight until three. After that I study, usually until seven or eight. Then HMD work: code reviews, team check-ins, decisions that need the chairman's input. Weekends are longer study sessions in the morning, deeper company work in the afternoon.
That sounds manageable when I write it as a list. In practice, there are days when a production incident lands in the middle of a physics problem set. Days when a difficult integral follows me into a code review and I catch myself thinking about substitution methods while reading TypeScript. Days when I finish both shifts and cannot remember what I accomplished in either.
The schedule works, not because I discovered some secret to doing two impossible things at once, but for a more boring reason: the company does not need me for most of what it does.
HMD Developments has engineers who ship features and handle incidents without my involvement. The other divisions have people running day-to-day operations. My role is direction, not execution. If I had to write every line of code myself, the whole arrangement would have collapsed months ago.
People assume that running a company and studying for an exam are completely different activities. That switching between them costs something extraordinary.
They are not as different as you would think. Debugging code and solving physics problems use the same underlying skill: decompose the problem, form a hypothesis, test it, adjust. The notation changes. The thinking does not. Some of the best problem-solving intuition I bring to software comes from mathematics, and some of my clearest thinking about mathematics happens after a long code review, when my brain is already in "break this down" mode.
That said. There are genuine costs.
Social life is the first casualty. My friends spend their free time together. I spend mine alternating between textbooks and terminals. I have accepted this, but accepting something does not mean you stop noticing it.
Guilt is the second. When I study, some part of my brain counts the unreviewed pull requests. When I work on HMD, some part calculates how many practice problems I am not solving. This background noise never fully stops. Managing it is its own skill, and I would not say I have mastered it.
Sleep is the third, on bad weeks. I am strict about this now because I learned the hard way that a tired brain is slow at both chemistry and code. Eight hours is not optional. It is the foundation everything else sits on.
The YKS will come and go. One day, one exam, one score. HMD Corporation existed before it and will exist after it.
But the ability to hold two large, competing responsibilities in your head simultaneously, to switch between them dozens of times a day without losing quality in either, that is a skill I plan to keep using long after the trigonometry homework stops mattering.
I did finish it, by the way. At midnight. Then I reviewed the pull requests. Then I went to sleep.
The chemistry mock went fine.