Walk into any room at Kaz Software's Dhaka office, and you won't be able to see what's on an employee's monitor.
This isn't an accident of furniture or office design. It has been since Kaz’s first year.
Wahid Choudhury, the founder, describes his own desk as an example: a wall sits behind him, and the only way to approach him is from the front. No one can walk up behind his chair and glance at his screen on the way past.
Every workstation in the Kaz head office in Dhaka is arranged the same way. It is, by his account, one of the oldest deliberate decisions in the company's history, older than most of the policies Kaz has, in a company that famously tries to have as few written policies as possible.
The reasoning in Wahid's words: "We wanted to communicate that we don't care what you are doing on your screen at this moment, you are on Facebook, watching videos, or coding. Because in the end, we have a trusting relationship, and we have trust that you will deliver the task. We have built this trust into our culture from day one."
This is a strange decision, and worth sitting with before moving to what it produced.
Most knowledge-work organizations, in fact, most any organization that cares about productivity, reach for the opposite instinct. Open floor plans designed for visibility, screen-monitoring software, time tracking, managers who walk the floor, and so on.
The implicit theory behind that instinct is that people default to slacking unless watched, and that visibility is what keeps work honest. The philosophy takes a pessimistic view of human capability and agency. That humans are unreliable. It essentially creates an environment of low trust and surveillance. The consequence of such a culture is that people don’t feel trusted. They feel watched. Spontaneity dies. Ownership dies. People don’t feel a bond with the organization. As the company measures its people, people also become measured in what they give to the people. The relationship becomes hyper-transactional.
Kaz built the literal opposite, architecture that makes monitoring impossible by design, on a different implicit theory: that visibility produces compliance, not commitment, and that compliance is a worse foundation for a software company than commitment is.
Wahid's case for why this works rests on a fairly simple behavioral claim, one he's held for over twenty years of running the company: "Luckily in the software world, it's easy to measure best efforts. You can see the deliveries. You are given a task to write code within a certain time, can you do it or not?"
In other words, the seating policy isn't really about screens at all; it's a bet that output is observable enough, on a long enough time horizon, that the organization doesn't need to monitor the process to know whether someone is performing.
What you give up by not watching the process, you get back by being honest about what you're actually trying to measure.
The follow-on claim is the more interesting one, because it's about reciprocity rather than measurement: "If you get trust from somebody, if you are happy with somebody, you will give your best to return that happiness. When we are trusted with responsibility, we try to grow to honor that trust. When you trust people, people tend to work harder to keep it."
This is the philosophy underneath the desk arrangement. The policy isn't a productivity hack; it's a relationship the company is choosing to enter into, on the theory that people who are extended trust have a reason to be worthy of it that surveillance can't manufacture. And it makes sense. In most instances, when you put trust in someone, they try to rise to that trust. The opposite is also true. In the esoteric sense, the world delivers us our expectations. When we expect people to cheat, that is what they will do, regardless of what we do. The opposite is also true.
Wahid is candid that this isn't free of risk, and that he knows what the standard objection sounds like, because he hears it constantly from peers running other companies: "The complaint I hear a lot from my colleagues outside of Kaz is that, 'Yes, the team is performing well, but you have to monitor the team closely, otherwise there will be problems.'"
His answer isn't that monitoring doesn't work; he doesn't make that claim. His answer is more interesting: that letting go of control is uncomfortable in a specific way that has nothing to do with whether it's effective. "Letting the control go might feel a bit scary, that maybe they are not doing anything," he says. The fear is real. He's just decided, across two decades and as the company scaled, that the fear doesn't track the actual outcome.
This single policy connects to other things Kaz does that look, on the surface, like separate decisions but share the same underlying logic.
The company doesn't issue a handbook. New hires sometimes ask for one and are told there isn't one, because "everything is unwritten."
Hierarchy is deliberately flattened: nobody is called "Sir," and even respectful deference to senior staff ("Bhais") is actively discouraged, to the point where seniors are expected to make self-deprecating jokes about their own irrelevance ("the older a person is, the more his brain starts dying," is a running joke seniors tell on themselves) specifically to lower the social distance between levels.
None of these is a separate culture initiative taken on over time. They're the same bet that trust and informality produce better software-team behavior than structure and hierarchy do.
What the policy produced, by Wahid's own account, is harder to verify than the policy itself, since it rests on his interpretation rather than a controlled comparison.
But the claim he makes is that Kaz has an unusual reputation in Bangladesh's tech industry for producing alumni who go on to found their own companies or take senior roles elsewhere, and that this is connected to, among other things, a culture where people "get that opportunity to think outside the box and explore their curiosities," the way he says he himself did in a relaxed, low-pressure school environment as a child.
He's explicit that he doesn't have data for this; it's a pattern he's noticed and believes, not a measured outcome.
A note on what to watch for in this case
The interesting tension in this case isn't whether the policy "worked". Wahid offers no controlled comparison, and you should be skeptical of any founder's account of his own culture's success. The interesting tension is between two competing theories of what makes monitoring necessary in knowledge work, and where each theory's bet actually pays off.
The first theory says: people default toward minimal effort unless observed, so observability is a precondition for good output.
The second theory, which underlies the Kaz seating policy, says: observability of process is a poor substitute for observability of output. And in domains like software, where output is measurable (a feature either ships or it doesn't, a bug either gets fixed or it doesn't), monitoring of process is not just unnecessary but actively counterproductive, because it signals distrust, and distrust is corrosive to the kind of voluntary extra effort knowledge work actually depends on.
Pay attention to the specific condition Wahid attaches to his claim: "Luckily in the software world, it's easy to measure best efforts." He is not claiming this generalizes to every kind of work. The case for trust-based management here is conditional on output being legible enough to make process-monitoring redundant.
As you read other cases on org design, watch for whether the domain in question has that same property, easily observable output, or whether it's a domain (sales with long cycles, research with uncertain payoffs, creative work with subjective quality) where output is genuinely harder to read, and where the calculus behind a policy like this one might not transfer.
