You Are Not as Productive as You Think You Are
You are not as productive as you think you are.
Let me explain.
The Illusion of the Desk
We spend hours at our desk, staring at a big screen, doing something.
We write prompts. We watch things happen. But we don't actually do as much as we used to.
I've noticed this in my day-to-day life. I spend hours and hours at my desk, and then my brain gets foggy. I don't feel like I've done much work — and then I realize I'm actually wasting time being at my desk instead of working.
Because now, in this new AI era, it's simpler than ever to delegate tasks and start enjoying your life, taking care of yourself.
The Scrolling Gap
How many times have you noticed this: you start a task, the agent begins running, and you just open X, YouTube, Instagram, or TikTok and scroll until the chat has finished?
You condense that time into your "productive" time, even though it isn't.
Just because you're sitting at your desk waiting for an agent to finish a task doesn't mean you were productive. In fact, you weren't. The AI was.
We Need to Redefine "Work"
This is a harsh truth, because I'm a workaholic. I like working.
But we need to redefine the word "work" right now, because we're not doing much work anymore. We can control our agents from our phones. For example, Synara will soon have an iOS app, so you'll be able to control every provider from your phone without being at your desk at all.
Two Hours On Beats Ten Hours Half-On
Here's the part that surprised me most: once I cut the dead time, I didn't get less done. I got more.
Because there's a huge difference between ten hours of being half-on and two to four hours of real, deep work.
Ten hours of half-on looks like this: you write a prompt, pick up the phone, scroll, glance back at the screen, lose the thread of what you were even building, re-read your own code to remember the plan, get distracted again. Your attention never fully lands anywhere. At the end of the day you're drained, and the actual output is thin.
Two to four hours of deep work looks completely different. Phone in another room. One problem in your head at a time. You're actually thinking — about the architecture, the edge cases, what you're really trying to build — instead of babysitting a progress indicator.
The irony is that agents make deep work more valuable, not less. The typing is no longer the bottleneck. The thinking is. Deciding what to build, how the pieces fit, what "done" actually means, catching the thing the agent got subtly wrong — that's the work now, and none of it survives constant context switching.
So compress it. Give the hard thinking a real block with no phone in it, hand off the execution, and then close the laptop. You'll ship more in four focused hours than in a whole day of drifting between the screen and your feed.
Start Noticing the Breaks
My suggestion: start noticing and acknowledging the breaks you take while the agent is running. Once you see them clearly, you can understand how to actually schedule your day.
And that's a good thing. Because you can now start:
- Going to the gym way more
- Reading more
- Taking more sun
- Spending more time with your friends
- Going out
- Working on yourself in a lot more ways
Instead of sitting at your desk, watching lines appear, and deluding yourself that you're being productive.
The Only Question That Matters
Here's the test I use now, and it cuts through everything:
If the agent had finished this while I was at the gym, would the result be any different?
If the answer is no — and it almost always is — then go to the gym.
Hours at a desk used to be a decent proxy for work, because nothing moved unless your hands moved. That's over. The work still happens; it just doesn't need you sitting there to witness it. Watching it happen adds nothing to the output. It only subtracts from your life.
So stop measuring the hours. Start measuring what you shipped, and what else you did with the day.
The goal was never to be at your desk. It was to build things — and then go live.
I'm documenting the full build-in-public journey on X/Twitter, including what I ship, what breaks, and what I learn.