This experiment began as an attempt to avoid a very specific kind of conversation.

If you are married, you probably know the type.

The end-of-month discussion about house help.

Who came when. Who left early. What counts as “on time”.

A conversation that never quite concludes, yet reliably returns the next month.

Naturally, I wrote down what I thought was the problem.

“Track attendance of my daughter’s nanny.”

That sentence lasted about five minutes.

When the problem turned out to be a solution

It did not take long to realise I had already jumped ahead.

That sentence was not a problem statement.

It was a solution.

Once I paused and asked what the actual problem was, the framing shifted—uncomfortably at first.

Attendance was not the core issue.

Tracking was not the goal.

In fact, it was not about time at all.

It was about repeated negotiation, trust, and the small frictions that build up when nothing is written down.

Seen this way, the actors became clearer.

There was me and my wife.

There was the nanny and my daughter.

And there was the household itself, full of routines, expectations, and unspoken agreements.

This was not a systems problem.

It was a relationship problem that happened to need a system.

Solving the wrong problem on purpose

Like most explorations, this one began with ideas that looked good on paper.

• Ledgers

• Biometrics.

• Cameras.

• Wi-Fi based presence detection.

Technically, these approaches worked.

Socially, they created new problems.

Wi-Fi detection was a good example. It seemed harmless at first. Presence without effort. No explicit action required.

But giving Wi-Fi access did more than enable detection. It changed behaviour.

Data limits disappeared. Nanny’s phone usage increased. At one point, I even had to rotate the password.

A small convenience had quietly become permission.

That was an important lesson.

Solutions do not just solve problems. They introduce new ones—often social and behavioural ones before technical ones.

Asking better questions

Instead of building anything, I slowed down and started asking questions.

• Who performs the daily action?

• Who benefits from the system?

• Who carries the friction when things go wrong?

• What behaviour does this system encourage over time?

Every clever idea collapsed under one of these questions.

Biometrics felt accurate, but wildly disproportionate for a household.

Background automation felt convenient, but slippery.

One constraint kept eliminating entire branches of solutions.

If the system can work without the human being consciously involved, it is probably doing too much.

That single sentence ruled out half the solution space.

The MVP that actually mattered

What I eventually landed on was almost underwhelming.

A static NFC tag.
A phone.
A deliberate tap.

No background tracking.

No inference.

No silent automation.

Presence had to be explicit.

Action had to be intentional.

This was not about building the smartest system.

It was about building the smallest meaningful one.

Once this version worked, something interesting happened. The questions stopped changing shape.

There were still improvements to make. UX refinements. Feedback to incorporate. I even keep a small journal of what feels off.

But I was no longer rethinking the system itself. I was refining it.

At one point, I sent my wife a short video of the rough proof of concept. A phone, a tap, and a spreadsheet quietly filling up.

Her response was simple: “You do know a thing or two about software.”

That felt like enough validation to keep going.

Where Gen AI quietly helped

Gen AI was not the solution here. It played three supporting roles.

It helped prototype quickly, without the overhead of context switching - Script Kid

It helped reason end to end about constraints and failure modes - Solution Architect

And it helped surface hidden assumptions I had skipped past too easily - PM

The real value was not speed alone.

It was the ability to explore without commitment.

Feedback over finish

Even after the MVP worked, the work did not stop.

Most feedback was not about correctness. It was about experience.

When does the system feel ready?

What feedback reassures instead of alarms?

How long can something take before it feels broken?

Most iteration went into copy, timing, and tiny signals.

That is when it clicked for me.

When humans are in the loop, UX is the system.

What I deliberately did not build

This was the most satisfying part of the exercise.

I did not add dashboards.

I did not add summaries or insights.

I did not add nudges or scoring.

Not because they were hard, but because they would have changed what the system was about.

There is a fine line between accountability and micromanagement.

Between presence and surveillance.

Gen AI makes it very easy to cross that line accidentally.

Closing thought

Not every system needs to be smart.

Some just need to be careful.

As Gen AI lowers the cost of building, the harder and more interesting work might be deciding what not to build.

Especially when humans are still very much in the loop.

PS: Explore the project on GitHub

This isn’t a post about minimalism, systems thinking, or conscious consumption.

It’s closer to a thank-you note.

Over the course of 2025, I found myself enjoying something unexpected. Certain products entered our home, solved very specific problems, and then quietly disappeared from conversation — either because they worked, or because they no longer needed to stay

As a product engineer, that disappearance felt impressive.

So this is me enjoying being a user who noticed good product design.

The air purifier

We bought an air purifier because a doctor recommended it. No research rabbit holes. Just a tired parent doing what was suggested.

What followed wasn’t dramatic, but it was real: fewer hospital visits, calmer nights during common colds, and less of that half-awake listening-for-breathing anxiety.

It didn’t fix everything.
It reduced worry.

That felt like good design.

The laptop organizer

At some point, two laptops and an iPad started living everywhere except where we expected them to be.

Desk. Dining table. Sofa arm. Occasionally the floor.

The problem wasn’t clutter — it was searching. Every morning came with a small scavenger hunt and a fear of something falling and breaking.

We bought a simple laptop organizer.

What changed wasn’t discipline. It was expectation.

There is now one place laptops belong. We don’t have to remember anymore. The organizer remembers for us.

The key holder (my personal favorite)

We have three keys that matter every day — bike, car, and house — and a few that mostly don’t. I am exceptionally bad at putting keys back where they belong.

Laundry pockets have been their retirement home.

We bought a magnetic key holder — an adhesive-mounted plate with a strong magnetic keyring. No hooks. No slots. Just magnet meets metal.

What surprised me wasn’t that it held the keys. It was how decisively it did so.

You don’t place the keys — you let go. The snap is tactile and audible. If the key isn’t there, it’s immediately obvious. Not five minutes before we leave. Immediately.

It doesn’t try to fix my habit.
It removes the moment where my habit usually fails.

A tiny product that designed around human forgetfulness instead of fighting it.

The dishwasher

Morning anxiety used to start with a question mark: Will the maid come today?

If the answer was no, the day began with vessels instead of breakfast, followed by rushed decisions and Swiggy guilt.

The dishwasher removed that entire branch of thinking.

As someone who manages people, I find it mildly ironic how comforting machines can be when predictability matters most. The dishwasher doesn’t negotiate. It doesn’t surprise you. It does exactly what it promised.

The science helps too — steam, temperature, detergents. It feels safer.

And feelings matter.

The sofa cover

Two kids meant food crumbs slowly migrating into sofa corners and turning furniture into places we didn’t want to inspect too closely.

We debated an expensive vacuum. Instead, we tried a free-size sofa cover.

It wasn’t elegant. It worked.

Crumbs stopped accumulating. Cleaning became easier. Pests reduced. Pressure reduced. Sometimes the simplest intervention buys the most peace.

Removing the TV. Adding a projector.

We intentionally removed the television from the hall.

Not as a parenting statement — but because it had quietly become the default. The kids stopped noticing the play park, the swimming pool, the cycles in the parking. Bored eating crept in.

We didn’t want to remove digital fun entirely, though.

So we bought a projector and a sound bar.

Now movies require setup. Waiting. Darkness. Intention.

Movie nights feel like events again. The kids love the experience — partly because it’s tech, partly because it isn’t always on. A reminder that constraints can improve enjoyment.

The car bed

Long drives to our native place had become exhausting. One child outgrew the back seat. The other struggled to nap.

We bought a car bed as an experiment.

Travel became calmer. Naps happened. Peace returned.

It’s not perfect — safety trade-offs still bother me — but as a prototype, it solved a real problem. Sometimes that’s enough to keep going.

The monitor

I used to run a dual-screen setup as a developer. While setting up my home office this time, I paused.

I’m not optimizing for code anymore. My days are meetings, reading, writing.

I chose a single large monitor instead — tightly integrated with my Mac.

Less visual noise. More calm.

Good design sometimes shows up as subtraction.

The Samsung tags

As travel picked up, a new kind of anxiety showed up — not losing things, but wondering if we might.

We added Samsung tags to a few predictable items.

What they really did wasn’t tracking. It was reassurance.

Knowing something could be found if needed lowered the background noise in my head. I stopped double-checking. Stopped replaying where did I last see it?

Good design sometimes works by giving you permission to stop worrying — even if you never actually use the feature you paid for.

The bookshelf

The bookshelf came in during yet another decluttering exercise — but this time with a slightly different intent.

It wasn’t about storage. It was about visibility.

We wanted reading to be obvious. Not aspirational. Not scheduled. Just present. In the line of sight. Easy to reach. Hard to ignore.

Slowly, it stopped being my bookshelf. It became ours.

Books moved in and out. Conversations started around them. It turned into a quiet symbol of shared curiosity at home — not enforced, just there.

Sometimes good design doesn’t create habits.
It creates conditions.

Subscriptions (the quietest products)

Subscriptions taught me something different.

They don’t fail loudly. They just fade.

Some stayed because they earned their place.
Lingo Kids works because my younger one enjoys it and keeps coming back. Learning happens without announcing itself. Importantly we feel in control

ChatGPT Pro stayed. Claude didn’t. Once canvas launched in GPT, the decision made itself.

YouTube Premium has been quietly invaluable with two kids at home. Netflix and Prime continue because the value equation hasn’t changed. Hotstar stays mostly for my mom — and occasionally feeds my cricket brain.

Others left without drama.

A weekly magazine never made it into our reading loop. A system design subscription left when I stopped needing it. Sun NXT saw no traffic. FirstCry ended when diapers did.

Internet was more decisive. No number of upgrades fixed Airtel’s reliability for us. Switching to ACT did — stable speeds and half-yearly billing. Fewer reminders. Fewer interruptions. Easily one of the best decisions of the year.

Even groceries mattered. We moved milk delivery from BigBasket to Akshaya Kalpa — not for price, but for hygiene and last-mile care. Covered delivery. Thoughtful handling. Small details that build trust.

Google One stayed because it never asked for attention. Airtel mobile stayed because spam protection improved and roaming became frictionless.

The best subscriptions were the ones that asked the least from us.

Looking back, this still isn’t a list of purchases.

It’s a record of moments where good design quietly reduced effort, anxiety, or friction — and then got out of the way.

One thing that stood out, though, was how differently retention showed up. Some products stayed because they kept raising the bar end to end. Others stayed simply because they continued doing their primary job well. And retention broke quickly once that core expectation slipped — improvements elsewhere stopped mattering.

As a user, it felt uncomplicated. Either the product keeps doing its main job and keeps getting better, or something else eventually replaces it.

If you’re curious about any specific product mentioned here, feel free to ask in the comments. Happy to share details.

Over the last few months, a pattern has repeated often enough that it no longer feels coincidental.

A senior front-end engineer, recently promoted, wonders aloud whether staying “just FE” is quietly narrowing his future. There are more interesting things happening elsewhere—platform work, AI—and he’s unsure whether he’s missing out.

A lead engineer, new to the team and performing well, talks about moving again. Not because the work is bad, but because he doesn’t want to spend the next few years doing “more of the same,” even if the same is going well.

A back-end engineer, comfortable, effective, recently promoted, mentions planning a move anyway. Nothing is wrong. He just doesn’t want to stay too long.

None of these conversations are new to someone who’s been managing engineers for close to a decade. What is new is the profile of the people bringing them up.

They’re high performers. Recently promoted. Engaged.
And the frequency has increased.

It made me pause—not to diagnose individuals, but to question my own assumptions about motivation.

For years, I’ve carried a fairly traditional model: growth happens through scope, recognition, compensation, trajectory. When someone feels restless, one of those levers usually helps. Lately, they don’t.

So I asked myself a quieter question:
Is something changing at a macro level that I’m failing to notice?

I spoke with my mentor. The signal I received was simple and unsettling: yes. Traditional levers—money, titles, long-term plans—don’t work the way they used to. Not because people don’t want them, but because waiting feels different now.

Waiting doesn’t feel neutral anymore

Most of the engineers I speak to aren’t anxious about survival. They’re anxious about relevance. About optionality collapsing faster than experience compounds. About timelines shrinking in ways no career ladder fully acknowledges.

AI is part of that background pressure—not always as an explicit threat, but as a constant uncertainty. So is visibility. Everyone can see faster paths, adjacent paths, improbable outcomes that nonetheless feel nearby.

When patience no longer feels neutral, restlessness starts to look rational.

This doesn’t feel like disengagement in the classic sense. No one is checking out. If anything, they’re too aware.

The pressure seems to come from higher aspirations meeting constrained paths. People want judgment, leverage, voice. They want to feel that their decisions matter—not eventually, but now.

And when the traditional signals of progress are slow, opaque, or misaligned with how fast the world appears to be moving, people look for motion elsewhere. Switching domains. Switching teams. Switching narratives.

Sometimes those moves help. Sometimes they just provide temporary relief.

Agency as the missing variable

What helped me make sense of this was a reframing I came across while thinking through these conversations: agency over intelligence.

Not skill. Not output. Agency—the felt ability to shape one’s trajectory.

Many of the engineers I work with are exceptionally capable. But capability without perceived agency creates friction. You can be good and still feel stuck. You can be promoted and still feel late.

From that lens, some career decisions that look impulsive begin to look like attempts to reclaim control—even if they carry real risk.

The manager’s tension

This leaves me in an uncomfortable middle.

As an engineering manager, I’m responsible for creating conditions where people can do meaningful work and grow. Historically, retention followed naturally from that.

Lately, retention feels less coupled to intent or quality. Even strong environments don’t fully offset the broader forces at play.

It leaves me with an unresolved question:

What’s going to change in the way we assess & compensate developers that retention becomes relevant?

I’m not sure whether that question points to an existential crisis of my role—or an opportunity to lead differently.

What compounds quietly

If agency is the limiting nutrient, it’s also the hardest to “provide.”

Beyond a certain point, agency doesn’t come from tickets closed or systems mastered. It comes from judgment, influence, and context. From being trusted to shape direction, not just execute within it.

These are slow skills. They resist clean benchmarks. They don’t map neatly to performance frameworks designed decades ago. Yet they compound faster than most technical advantages once they take hold.

The irony is that many of the engineers most restless today are the ones best positioned to develop these capabilities—if the system made room for that kind of growth.

An unfinished thought

This isn’t happening everywhere. Not yet. But it’s disproportionately visible among people you’d put on your A-team.

I don’t have a clean conclusion. Only a sense that something fundamental is shifting—away from patience as virtue, toward agency as survival.

For now, I’m trying to listen more carefully—to what’s being asked beneath the movement.

Notes

• A framing that helped me articulate agency came from a short post by Andrej Karpathy, contrasting agency with intelligence as a scarce and powerful trait.

• A longer thread by systematicls offered a broader socioeconomic lens on why patience feels less rational under compressed timelines and structural uncertainty.

These didn’t provide answers—but they helped me name the pressure more clearly.

In October 2024, I joined the Gryffindor team as an Engineering Manager. Around the same time, Atlassian’s biannual Employee Wellbeing Survey results were shared.

One question stood out:

Do you feel connected as a team?

Roughly one in three people felt this was an area that needed serious improvement.

There were reasonable explanations. The team had operated without a direct manager for over a quarter. Nearly 30% of the team had less than three months of tenure. In many ways, the team was still forming—while delivering.

What surprised me wasn’t the score itself, but the realization that followed.

Until the survey, no one had articulated this as a problem. Work was moving. Rituals existed. Meetings happened. It was only when people were asked to name their experience that a shared awareness surfaced.

Disconnected didn’t look broken.
It looked… quiet.

Two problems, not one

Except for a brief stretch during COVID—when everything felt provisional—I had mostly managed co-located teams.

So I was facing two things at once:

1. A team that didn’t feel sufficiently connected

2. A 100% remote setup, where many of my instincts were untested

I didn’t have a playbook, and I was cautious about importing one.

Instead of starting with solutions, I started with a question:

What does connectedness look like when proximity is removed—and what small things does a system need to make it easier?

A working lens

One idea stayed with me: belonging often comes from shared context, not forced interaction.

In remote teams, many low-effort ways of discovering common ground disappear. You don’t overhear conversations. You don’t walk out of meetings together. You don’t casually learn what people care about.

Over time, my hypothesis simplified:

In remote environments, connectedness erodes not because people don’t care—but because the system removes easy ways of caring.

Small experiments in shared context

One of the earliest conversations we had was about companionship.

Not friendship.
Not closeness.

Just the feeling that someone is alongside you while the work happens.

With that in mind, we tried something small.

A Secret Santa, reframed as “Know Your Peer.” Each person was randomly paired with a teammate, with one condition: the gift had to come from understanding the other person’s interests. We quietly announced a small prize for the most thoughtful gift—not the most expensive one.

Alongside this, we kept a recurring, optional Monday prompt: “How was your weekend?”
And a few lightweight discussion spaces—interest groups people could drift in and out of.

None of this was mandatory.
None of it was positioned as culture work.

They were simply invitations. In hindsight, these moments made it easier to notice where shared interests already existed—something we leaned into later, asynchronously.

Time that was explicitly not productive

Remote teams are efficient by default. Over time, that can signal that presence must justify itself with an agenda.

We experimented with no-agenda, fun-only sessions. Not offsites. Not quarterly events. Just time.

Games changed—Among Us, Pictionary, meme wars, even investing simulations. The specifics mattered less than the signal.

I was careful not to delegate this away or attend selectively. Systems learn from attention. When “fun” is optional, it becomes expendable.

Letting go of synchrony

One constraint I had underestimated was presence.

In remote setups, synchrony can quietly become the price of inclusion. If you miss the moment, you miss the connection.

So over time, the team leaned into asynchronous participation:

Interest groups — spaces where people could engage with peers who shared a common interest, whenever they wanted, without expectation or cadence.
Social threads during festivals, work anniversaries, and birthdays — no commentary required.

Lowering the cost of participation mattered more than increasing frequency.

Restraint as a design choice

Scale was tempting—but fragile.

Instead, we kept things small and repeatable.

Two principles guided us:

• It should always be okay to skip

• Those who do participate should leave feeling lighter, not obligated

Every so often, I reminded the team why we were being intentional—not to enforce consistency, but to make the intention visible.

Connectedness doesn’t sustain itself silently.

My role in all this

I didn’t build connectedness. I didn’t drive culture.

At best, I was a participant and a catalyst.

What made things memorable was the team—their willingness to show up, sometimes awkwardly, sometimes playfully, but genuinely. The shared moments, inside jokes, and memories were created by the team, not by design.

I simply helped create conditions where those moments were easier to happen.

No conclusions (yet)

In the two biannual surveys that followed, Gryffindor scored 100% on the connectedness question.

I’m cautious about what to make of that.

Surveys are snapshots.
Teams mature.
People settle in.

I don’t read this as proof that anything worked. At best, it suggests the environment shifted—enough for people to answer a question differently.

What feels more durable than the score is harder to measure.

People now reference shared moments without explanation.
There are inside jokes that don’t need context.
Names come with stories, not just roles.

We know each other a little more as humans.

Those memories will likely outlive our Atlassian avatars—and that feels like the right place to pause.

Reading lineage

A few ideas in this post were shaped by earlier reading. I’m listing them here as context, not explanation.

• First, Break All the Rules

• Harvard Business Review Emotional Intelligence Series

These aren’t playbooks. If anything, they nudged me to notice how often connection emerges from conditions, not instruction.

We played housie.

Nothing fancy. Paper tickets, a caller with a mic, and a room full of people pretending not to care too much. It was at a house party hosted by us — an evening meant purely for entertainment. No agenda. No outcomes. Just a game to fill the space between conversations.

At first, it was loud. Numbers being called out. Laughter. The occasional groan when a number was missed by one.

But after a few rounds, the room changed.

Some people grew quieter. Some stopped checking their tickets altogether. A few celebrated their wins, but almost apologetically — as if unsure whether it was okay to feel happy while others were still waiting.

I wasn’t playing as much as I was watching. As the caller, I had the vantage point to observe what followed every number call and prize announcement. Even in a game where no one could influence another person’s ticket, competitiveness surfaced. Not overtly, but unmistakably.

That’s when it struck me: I had seen this before.

Not during a game.
During the work itself — when outcomes mattered and time was limited.

Over time, I had begun to notice something similar unfolding there. People weren’t just focused on delivering work. They were also protecting themselves.

Not in obvious ways. No one said it out loud. But it showed up in subtle, human responses. Conversations became cautious. Feedback loops tightened. Neutral observations felt heavier than intended.

When delivery pressure is high, judgment — real or imagined — arrives early. Even well-intended input can feel like an attempt to downplay effort. Listening becomes selective. Trust thins, quietly.

Competition doesn’t announce itself in these moments — it creeps in, shaping how people listen and respond.

I started forming a hypothesis.

When people pursue similar outcomes in a constrained system — shared goals, limited time, visible impact — two walls tend to form involuntarily.

The first wall is built to avoid conflict. Conflict risks friction, and friction risks slowing momentum. So disagreement is softened, opinions are withheld, and silence starts to feel safer.

The second wall is built to avoid comparison. Not wanting to feel behind a peer, responses tend to oscillate between defensiveness and withdrawal.

Neither reaction felt irrational. In fact, both felt deeply human.

What the housie game clarified for me was this: even when outcomes are purely random, comparison still finds a way in — not because people want to compete, but because relative position quietly matters.

If competition can surface in a game of chance, it’s not surprising that it shows up more strongly in day-to-day work, where effort, identity, and outcome are tightly coupled.

The problem isn’t competition itself. That part is inevitable.
The problem begins when we forget what kind of game we are actually playing.

Much of our work is meant to be collaborative — a process of learning, iterating, and getting better together while moving toward outcomes.

Calibration, at its core, is meant to support that journey — to align on expectations, recognize progress, and reinforce behaviors that help the team and the business grow. Not to rank people against one another, but to set a shared bar for mastery.

When the system feels competitive, teams don’t stop working hard — they stop learning together.

That learning depends on a basic trust — trust in the competence of the people around us, and trust that the process will be fair even when outcomes differ. When that trust weakens, comparison takes over, and protection feels safer than openness.

Connectedness becomes the quiet casualty. Not because people don’t care, but because self-protection starts to feel necessary.

What stayed with me was not whether this parallel landed for everyone. It may have. It may not have.

What stayed with me was how easily competition slips into places it doesn’t belong.

Once you notice that, you start listening differently — not just to what people say, but to the pressures shaping what gets said, and what doesn’t.

And maybe that awareness itself is the beginning.

Not of eliminating competition — that would be unrealistic — but of remembering that this was always meant to be a team game.

Some games end with a prize.
Others end by revealing something about us.