Forem: Mr Chandravanshi

AI as a Job Threat: The Hype Is Bigger Than the Reality

Mr Chandravanshi — Sun, 19 Apr 2026 08:38:00 +0000

The Job That Stayed. The Work That Left.

AI is not removing positions from Indian workplaces. It is removing what those positions used to build inside the people holding them.

Peter approves thirty-six rows before noon.

The software read twelve hundred invoices. The software matched the vendors. The software entered the numbers. It flagged the rows where its confidence fell below 97 per cent. He checks those, approves or escalates, and closes the laptop.

His mother thinks he got promoted.

He told her he became a babysitter for a robot. She didn't hear the second part.

The Job That Stopped Building Anything

For three years before the software arrived, he read those invoices himself. Messy formats, faded ink, rubber stamps pressed at angles. He got fast. He learned which vendors rounded up, which tax codes carried habitual errors, which gaps were typing mistakes and which were real problems.

Ten hours a day. Thousands of documents. His hands moved before his thinking did.

That knowledge is not being built anymore.

Not because he lost the job. Because the job stopped requiring it.

What replaced it is narrower: catch what the software missed. Three per cent of twelve hundred items are reviewed each morning. The function does not deepen with repetition. It does not transfer to another role. Nobody designed it deliberately. It appeared when the previous role was removed, and something had to occupy the gap.

The title still says Accounts Executive.

Three Positions Inside the Same Shift

His father, fifty-four, learned Excel three years ago when his factory in Bihar moved from paper ledgers to spreadsheets. Two choices: learn or leave. He learned. Four months, help from his nephew, several evenings, angry at a machine that seemed to resist him personally.

He got through it. Still employed. Still useful.

That transition cost something real and produced something real. A skill with a destination. The embarrassment of being slow in front of younger colleagues turned into competence that paid him back.

What Peter Wilkinson is doing now has no comparable destination.

Near the Patna market, an eighteen-year-old started six months ago making logos and posts for kirana stores. YouTube. AI tools. Three hundred rupees per logo. Twenty clients this month.

He didn't lose a job to AI. He entered the market because AI lowered the entry requirements.

Same shift. Three completely different positions inside it.

The variable isn't effort or ability. It is where each person stood in building a skill when the tool arrived. Peter was deep inside one. His father had to rebuild toward another. The shop kid had nothing to lose, so he built on the new terms without grief.

What the Speed Removed

At a consultancy in South Delhi, junior analysts stopped writing first drafts sometime in the past two years. They approve them now. Forty reports a day. The output numbers look extraordinary.

When she used to build a quarterly report over two days, she noticed things while compiling. A vendor whose costs had risen three quarters running. A category that underperformed in the same month every year. That noticing wasn't in her job description. It was a side effect of doing the work slowly, by hand, with full attention on every cell.

The forty-minute version produces a cleaner document.

It does not produce that side effect.

She is twenty-three. The gap is invisible now. It will show up later, in a room where someone needs to build something from nothing, and she will find that two years of approving other people's arguments did not prepare her for it.

Reading is not the same as writing. Approving is not the same as building.

Who Is Talking and Who Is Living It

The loudest voices in the AI-and-jobs debate belong to people whose work involves talking about work. Tech newsletters. Conference panels. LinkedIn posts written by people whose job is writing LinkedIn posts.

Their work is not being hollowed. If anything, the volume of things to say about AI is expanding what they produce.

The people actually inside the shift are not being quoted. They are on rooftops in Indore reviewing flagged rows. In South Delhi offices, signing off on documents they didn't write. Fifty-four years old in Bihar, learning software because leaving is the only other option.

No announcement arrived. Nothing collapsed. The salary is the same.

What changed is what the day deposits inside the person living it.

The job used to turn effort into ability. Slowly, without announcing it, across thousands of repetitions. That is no longer happening, and it does not appear in any employment count, any productivity report, or any labour survey currently running in India.

Closing

Peter closes the laptop. His mother brings tea and asks if the work is getting better.

He says yes.

The job is there. The thing the job used to do to him is not. He cannot explain the difference in terms she would know how to respond to.

Nobody can yet.

AI Isn’t Taking Jobs. It’s Revealing Which Ones Were Fragile

Mr Chandravanshi — Sun, 19 Apr 2026 05:24:00 +0000

What looks like disruption is often a slow exposure of work that was always easy to replace

The fear around AI focuses on loss.

Jobs disappearing. Roles becoming irrelevant. Entire categories of work fading out.

That reading feels intuitive.

It also misses what is actually happening.

What AI is actually doing

AI is not starting the change.

It is revealing where the change was already possible.

For a long time, certain kinds of work felt stable.

Clear instructions. Defined steps. Predictable output.

You follow a process and get a result.

That structure creates a sense of security.

It also creates a hidden weakness.

Anything that can be described completely can be transferred.

First to another person.

Then to a tool.

And once that transfer becomes easier, the role begins to shift.

Not at the level of the job title.

Inside the work itself.

Where the shift begins

One task becomes faster.

Another requires less attention.

A third is no longer needed in the same way.

The role continues to exist.

But the work inside it starts changing shape.

This is where most people misread the situation.

They look for a clear break.

A moment where the job disappears.

Instead, the change spreads through smaller parts.

Task by task
Step by step

By the time it becomes visible at the role level, the shift has already happened.

This pattern isn’t new

In the early 1990s, when computers entered offices, the visible impact looked similar.

Manual processes reduced.

Clerical work declined.

From the outside, it appeared as removal.

At the same time, new forms of work were expanding.

Software development
IT support
System maintenance

These roles were not replacing old ones directly.

They formed around what the technology made possible.

The internet repeated it

The same thing happened in the early 2000s with the internet.

Work stopped depending on a fixed location in the same way.

People began working remotely.

New kinds of digital roles appeared.

Again, the visible change was reduction in one place.

The actual shift was movement.

Why AI feels different

AI is following that same pattern.

But it feels more uncomfortable because it reaches into cognitive work.

Writing, analysis, decision support.

Areas people assumed would remain stable for longer.

The underlying process remains the same.

What gets affected first

Work that follows clear steps changes first.

Repetitive tasks.

Defined workflows.

Predictable outcomes.

That work is easier to transfer.

So it gets compressed.

Compression reduces how much human effort is required.

Fewer people are needed for the same output.

That looks like job loss.

It is closer to job filtering.

Some work holds.

Some work does not.

What actually determines survival

The difference is not effort.

It is how replaceable the work was from the beginning.

The constraint most people ignore

There is also a limit that shapes this shift.

If income disappears at scale, demand disappears with it.

Without demand, businesses cannot function.

Without functioning businesses, the system breaks.

This does not prevent disruption.

But it prevents total removal.

The system adjusts because it has to.

How work reorganizes

Work reduces in one place.

Expands in another.

New roles form around what becomes possible.

They do not arrive as direct replacements.

They appear in different forms.

Often without clear names in the early stages.

This is why people miss them.

They look for the same roles they understand.

The change happens somewhere else.

What separates people

The difference between people is not simply skill.

It is timing.

Some move with the shift.

Some stay with work that is becoming easier to replace.

The gap does not appear immediately.

It grows over time.

The real takeaway

AI is not introducing a new outcome.

It is accelerating a familiar one.

Work does not disappear in a clean line.

It reorganizes around what is easier to automate and what still requires human judgment.

The part that feels sudden is not the change itself.

It is the moment people realize the work they relied on was already moving away from them.

AI Won’t Kill Jobs. It Will Reshape Work Itself

Mr Chandravanshi — Sun, 19 Apr 2026 04:50:42 +0000

AI Won’t Kill Jobs. It Will Change What “Work” Means

The fear feels new. The pattern is not.

The fear around AI sounds absolute.

Jobs will disappear. Entire careers will collapse. People won’t be needed in the same way.

It feels like a clean break from the past.

It isn’t.

This belief assumes something that has rarely been true. That work disappears in a straight line when technology improves. That replacement is direct. That what exists today simply vanishes tomorrow.

That is not how it has worked.

What history actually shows

When computers entered offices in the late 1980s and early 1990s, the visible change looked similar. Clerical roles reduced. Manual record-keeping faded. Tasks that once required hours were completed in minutes.

From the outside, it looked like removal.

But that reading missed what was forming at the same time.

Software development expanded. IT services grew. Maintenance, training, and system management became work in their own right. Entire industries formed around the machines that were supposed to eliminate work.

The same pattern repeated when the internet spread in the early 2000s.

Location stopped being a strict boundary. Work moved into homes. A laptop and a connection became enough for income in many cases. Roles changed shape. Some disappeared. Others appeared that did not exist a decade earlier.

The common mistake in both moments was treating visible loss as the full story.

What is actually changing with AI

AI is being read through that same narrow frame.

What is visible now is the pressure on repetitive work. Tasks with clear steps. Processes with fixed inputs and predictable outputs. These are the easiest to compress.

Compression is often mistaken for disappearance.

But they are not the same.

When a task is compressed, it requires less human time. Less repetition. Fewer people for the same output. The work has not vanished. It has been absorbed into tools.

That absorption creates space.

Not immediately visible space. Not cleanly defined roles. But capacity that did not exist before.

Where new work comes from

The difficulty is that new work does not arrive labeled as a replacement for old work.

It forms around what the technology makes easier.

After computers, it formed around managing and building software.

After the internet, it formed around distribution, communication, and digital services.

With AI, it is forming around judgment, coordination, and the ability to work with systems that no longer need step-by-step instruction.

Why this shift feels different

This shift is harder to see because it does not announce itself.

It does not remove a job title overnight.

It changes the weight of the work inside it.

A task disappears here. Another becomes lighter there. A responsibility shifts toward oversight instead of execution. The role still exists, but it does not feel the same.

This is where most people misread the situation.

They look for disappearance at the level of jobs.

The change happens at the level of tasks.

And tasks move faster than titles.

The constraint most people ignore

There is also a constraint that rarely enters the conversation.

If income disappears at scale, demand disappears with it. No demand means no functioning market. No functioning market means the system cannot sustain itself.

This does not prevent disruption.

But it limits total collapse.

Work does not vanish completely because the system requires participation to function.

How the system adjusts

So the adjustment happens.

Not as a coordinated plan.

But as a response to pressure.

Roles shrink in one place
They expand in another
New forms of work appear around what becomes possible, not around what was lost

This process is uneven. It does not happen at the same speed for everyone.

Some people move early, often by accident. Others stay anchored to work that is quietly losing relevance.

The difference does not appear immediately.

It shows over time.

What this actually means

AI is not introducing a new outcome.

It is accelerating a familiar one.

Work is not ending.

It is moving away from what can be reduced to steps, and toward what cannot.

The part that feels sudden is usually the moment people notice, not the moment the change began.

The Fed's Three Levers

Mr Chandravanshi — Sat, 18 Apr 2026 19:53:19 +0000

How Language, Discount Rates, and Balance Sheet Operations Actually Move Stock Prices

The Federal Reserve has three instruments that move equity markets. Only one gets reported.

The rate decision — the number announced after each Federal Open Market Committee meeting — is the one that appears in headlines. It is also the one that does the least real-time work.

By the time the Fed announces a rate change, institutional markets have already priced it. What moves prices is what comes before the decision: the language that shapes what participants believe the decision will be.

Understanding all three instruments explains almost everything that gets attributed to vague forces like "market uncertainty" or "Fed fear."

Lever One: The Sentence Before the Action

In September 2022, Jerome Powell spoke for eight minutes at Jackson Hole, Wyoming. No new data. No surprise announcement.

He said what analysts already knew — inflation was high, rates would stay elevated, the Fed would act. Markets fell 3.4% that session. The following week erased trillions in capitalization.

The rate hadn't changed. The sentence had.

Forward guidance is the practice of communicating future policy intentions publicly. Its power runs through a specific channel: current decisions are shaped by what people believe will happen next.

A company deciding whether to expand does not look at today's borrowing costs — it prices off the expected rate twelve months out.

A household taking a variable-rate mortgage is making a bet on the future rate path. When a central bank can reliably shape that forecast, it changes present behavior without changing anything present.

The statement does the work the rate decision would have done — weeks earlier, at no cost.

This is why institutional desks read Fed statements word by word against previous ones.

Not paragraph by paragraph. Word by word. "Patient" replaced by "data-dependent" is not editing. It is a signal that hundreds of billions in positioning will reprice within hours.

A single phrase — "we expect to maintain an accommodative stance until maximum employment is achieved" — carries three distinct embedded signals: current policy is accommodative; the benchmark for change is employment, not inflation; the timeline is open-ended.

Three different market segments move on three different parts of that sentence.

The Fed adopted this instrument gradually. Through the 1970s and most of the 1980s, the Federal Open Market Committee said nothing publicly for weeks after each meeting.

Markets learned policy from actions, not words. The shift accelerated under Ben Bernanke, who came to the chairmanship with academic work on central bank communication and a conviction that transparency improved policy effectiveness.

The 2008 collapse made forward guidance structurally necessary rather than optional.

When the Fed cut rates to zero in December 2008, the conventional lever was gone — rates cannot fall below zero without creating incentives to withdraw deposits entirely rather than pay to store them.

By committing to hold at zero "for an extended period," and later through calendar-based and threshold-based commitments, the Fed pushed down longer-term rates even when short-term rates could not move.

If markets believed the short rate would hold at zero for two years, two-year bond yields fell toward zero without any direct Fed action on those securities. The words manufactured the effect.

Lever Two: The Discount Rate and Every Stock Price

The connection between Fed rate decisions and individual equity prices runs through arithmetic, not sentiment.

A stock price is the sum of all expected future earnings, discounted back to what they are worth today. The discount rate is the variable.

Near zero, a dollar of earnings ten years from now is worth nearly a dollar today. At 4 or 5%, that same future dollar is worth considerably less. The price falls — not because the business deteriorated, but because the calculation changed.

High-growth companies sit at maximum exposure here. Their value is concentrated in distant projected earnings, not current cash flows.

In 2022, Apple, Amazon, and Tesla each fell 40 to 50% over the course of the year. Their businesses did not collapse. Revenue held. Products sold. What changed was the rate applied to every future dollar of earnings — and the math followed automatically.

The Fed sets short-term rates directly. Long-term rates — ten-year Treasuries, corporate bonds, mortgages — are set by market participants aggregating their beliefs about future inflation, growth, and policy.

The Fed cannot dictate these. It shapes them through the expectation channel described above.

The full transmission runs:
*Fed language → market belief about future rates → long-term rate level → discount rate applied to equities → stock prices. *

The announcement at the end of that chain is the last event in a sequence that began weeks earlier.

Lever Three: Where the Balance Sheet Enters

Quantitative easing operates through a different channel entirely — one that connects the Fed's balance sheet directly to asset prices without passing through the rate mechanism.

When the Fed buys Treasury bonds or mortgage-backed securities, two things happen simultaneously. Sellers receive cash they must redeploy it somewhere.

The supply of safe assets available to private investors contracts. Both effects push capital toward riskier alternatives: equities, credit markets, and real estate.

Prices rise not because earnings improved but because the alternative — holding cash or low-yielding bonds — became comparatively unattractive.

By early 2022, the Fed's balance sheet had reached nearly $9 trillion — more than double its pre-pandemic size, built through monthly purchases of $120 billion in securities through much of the pandemic period.

That buying had displaced enormous quantities of capital into risk assets over several years.

When the Fed reversed direction through quantitative tightening — allowing holdings to mature without reinvestment — that liquidity came back out.

Smaller balance sheet, less capital seeking returns, and reduced upward pressure on asset prices. By some analyses, this effect on equity valuations equals or exceeds the effect of the rate level itself.

The 2022 tightening cycle combined all three levers moving in the same direction simultaneously: guidance signalling sustained elevation, rates rising 400 basis points in twelve months, and a $9 trillion balance sheet beginning to contract.

The result was one of the worst years for both stocks and bonds in modern market history — not because anything broke operationally, but because the conditions determining what everything was worth changed sharply and at once.

Where the System Fails

Forward guidance works when the Fed's description of the future matches the future it delivers. When the gap widens, the damage to credibility — and to markets — is proportional.

In early 2021, the Fed described rising inflation as "transitory." The word carried a specific signal: no aggressive response coming, price increases are temporary, the cycle continues.

Markets are positioned accordingly. By late 2021, inflation was persistent, broadening, and running at multi-decade highs.

The Fed abandoned not just a forecast but the framework built around it. Seven rate increases followed in 2022 — the fastest tightening cycle since the 1980s.

The losses were not only from the rate increases themselves. They were from the gap between what the guidance had signalled and what the policy actually delivered.

Powell's framing acknowledged the deliberateness: the tightening was more aggressive than conditions strictly required.

The excess was a public signal that the institution was serious about price stability, even at the cost of growth. Credibility lost through a wrong framework gets rebuilt through overaction in view of the market.

The Fed's quarterly Dot Plot institutionalised the forward signalling into a published document. Each dot represents one committee member's projection for where the federal funds rate will sit at year-end for each of the next several years.

Between September and December 2022, the distribution shifted visibly upward — more members saw rates staying higher for longer. Markets repriced before a single additional increase was announced. The dot shift was the event. The rate decision confirmed it.

The Dot Plot is a projection, not a commitment. It is revised quarterly. The gap between dots and outcomes is where most losses attributed to "Fed policy" actually originate — not from the decisions themselves, but from positioning built on projections that later changed.

What the Three Levers Produce Together

Stock prices are not reflections of corporate earnings.

They are reflections of corporate earnings discounted at a rate substantially controlled by the Fed, in a market where available capital is substantially shaped by the Fed's balance sheet, under conditions where every future decision is partially pre-priced through the Fed's language.

The heuristic "don't fight the Fed" is logically grounded. When all three levers move in the same direction, the effect is total — every borrower, every leveraged position, every equity valuation simultaneously.

No individual decision overcomes that kind of coordinated, system-wide pressure.

Its limitation is equally specific: using it requires knowing what the Fed will actually do, not what it said last.

Signal and action have diverged often enough, with damage large enough, that mechanically trading Fed statements carries a poor track record.

The durable frame is simpler. Rising rates compress valuations whether or not businesses perform. Falling rates expand them whether or not they deserve it.

Balance sheet expansion displaces capital into risk assets; contraction withdraws it. Language moves all of this before any decision is announced.

The rate is the confirmation. The three levers did the work before anyone reported it.

Why Investors Exit Right Before Recovery

Mr Chandravanshi — Sun, 12 Apr 2026 17:11:07 +0000

The endurance depletion mechanism behind panic selling at market bottoms

Panic selling at market bottoms is not caused by emotional weakness. It is caused by a finite cognitive resource - endurance under sustained loss - reaching zero.

The investor who exits on the worst day usually understands, clearly and completely, why they should not.

They exit because understanding something and being able to hold it through eleven consecutive days of loss confirmation are two different capacities.

One is knowledge. The other is a resource that depletes.

This distinction is not semantic. It determines what the actual solution is.

The depletion mechanism

Behavioural finance research on loss aversion measures single-event pain. Losses feel approximately twice as painful as equivalent gains feel good.

What this research does not cleanly capture is what happens when the same loss confirms itself across eleven consecutive mornings.

Each red day does not bring new information. It is the same information repeating. The brain processes repetition as pattern confirmation.

The pattern reads: this does not stop. That reading is structurally wrong near the bottom of a fall - recoveries do not announce themselves, and the day the market turns is indistinguishable from any other day in the fall.

But the investor is no longer making a prediction by day eleven. They are ending an experience that has exceeded their capacity to continue it.

The depletion is not linear. Each day, the investor has less psychological distance from the previous one. Day three draws down the resource slightly.

Day eight draws it down faster. By day ten or eleven, the cognitive resource that keeps a position stable under discomfort is at or near zero.

The exit that follows is not fear in the acute sense. It is a threshold being crossed.

This threshold explains the clustering pattern that puzzles analysts. Panic selling concentrates at market bottoms rather than distributing evenly across a fall.

Individual investors deplete at different rates, but they are all accumulating the same daily confirmations from the same fall.

Depletion peaks across the investor population at roughly the same point. Social signals - a WhatsApp group, a financial news cycle, visible exits by others - then provide permission at the moment of collective peak depletion.

One investor posts that they sold. Others who have privately reached the threshold receive confirmation that leaving is acceptable.

Exits cluster. Selling pressure concentrates at the bottom.

The mechanism is not herd behaviour in the simple sense.

It is individual threshold crossings, reached independently, made collectively visible through social permission at the worst structural moment.

One caveat requires precision: this mechanism applies to investors holding sound positions - diversified funds, index instruments, and reasonable asset allocation.

An investor who entered an overleveraged position, or concentrated in a sector undergoing genuine structural deterioration, may have been correct to exit.

Endurance depletion and investment accuracy sometimes coincide. The problem at peak depletion is that the investor cannot reliably distinguish which situation they are in.

The cognitive resource that supports clear analysis is the same resource that has been drawn to near zero.

The re-entry failure follows directly from the exit. The investor who sells tells themselves they will return when things stabilise. Stabilise is undefined.

In practice, it means that investing no longer feels painful. That condition is met after significant recovery has already occurred.

The investor re-enters at a higher price than their exit, having realised the loss and missed the recovery move.

The question they ask afterwards - why didn't I sell earlier - searches for a point where the loss was smaller.

It does not search for the points where staying meant full recovery. That search requires cognitive resources that the aftermath does not supply.

The only intervention that works

Telling an investor to manage their emotions during a sustained fall is not a solution. It is a description of the problem restated as an instruction.

Emotions are not a switch. Awareness that day ten will be painful does not reduce the pain on day ten or replenish the endurance being drawn down on days one through nine.

The intervention must operate before the fall begins, because the fall is precisely when it cannot operate.

A pre-commitment device. A written rule established during a period of full cognitive capacity that removes the exit decision from the moment of peak depletion. The rule must be structural, not aspirational.

"I will stay calm" is an aspiration. "I will not execute any portfolio exit within 30 days of a new portfolio low" is a structural rule.

"I will discuss any sell decision with one named person before executing - not someone in the same investment group chat" is a structural rule.

A specific review date, written in advance, before which no action is taken, regardless of portfolio movement - this is a structural rule.

The specificity is not bureaucratic. It is functional. A vague intention to remain rational places the decision back inside the moment of peak depletion.

A written rule with a specific date and a specific condition moves the decision outside that moment entirely.

The investor who has this rule does not need greater emotional strength than the investor who does not.

They needed only to have made one decision before the fall began that made the exit decision structurally unavailable at the worst time.

The investor who exits at the bottom is not a worse investor than the one who holds.

They held the same position through the same fall with the same knowledge.

They simply did not install a structure that protected their decision from the moment when their decision-making capacity was guaranteed to be at its lowest.

That structure is entirely installable before the next fall begins.

The window to install it is now, not during the fall.

Why Retail Investors Sell at Exactly the Wrong Moment

Mr Chandravanshi — Sat, 04 Apr 2026 09:03:12 +0000

The Hour, Not the Market

Why investors don’t break at the bottom — they break at midnight

The trigger is not the market falling. It is the hour, the body, and three weeks of carrying it alone.

The phone was already face down on the bed before Nishant had made any decision he could name. That detail matters. Not the red number. Not the eleven months. The specific moment when the body resolves something the mind is still arguing about - and neither one notices until it is already over.

He had entered the position carefully. A diversified equity mutual fund, moderate risk profile, five-year horizon. He had read the right things. He understood, in principle, that markets move in cycles and that long-term investors do not react to short-term noise. He could have explained this to someone else that morning with complete confidence.

By 11:47 PM, the explanation was still present. It just had less weight than the number on the screen.

The Accumulation

The previous seven months had been straightforward. The fund climbed slowly, which felt like confirmation. Then it flattened across two months, which felt like patience being tested. Then it fell across three weeks in a way that felt categorically different from anything before - not like weather passing through, but like something structural, like a decision the market had made and forgotten to announce.

The earlier dips had been easy to hold. This one accumulated differently.

Every check of the app was a small event. The number was red. The body registered something. Nothing was done about it. The app was closed. An hour later, the same sequence. Twice a day, then three times, then without counting. Each check resolved nothing and added a small amount to a load that had no release valve.

By the time Nishant opened Zerodha at 11:47 PM on a Tuesday in November, he had run that cycle dozens of times. The stress load he was carrying was not the product of that night. That night was just when the container ran out of room.

His wife was asleep. The fan was running. The room was quiet in the specific way that makes a temporary feeling seem like the only reasonable conclusion. The sell screen was two taps away.

What Behavioural Finance Misses

He had been managing the accumulation the way most retail investors manage it. Reading threads on Reddit where half the comments said hold, and the other half said the recovery was not coming. Repeating the long-term investing principle to himself - genuinely, not performatively, because he still believed it. Checking financial news for context that would make the number make sense.

None of it discharged the load. It just gave the load somewhere to look while building further.

This is the part behavioural finance explains correctly but incompletely: loss aversion. Humans feel losses roughly twice as intensely as equivalent gains. The brain processes financial threat similarly to physical threat. Cortisol rises. The chest tightens. Rational evaluation narrows.

What behavioural finance does not explain fully is the timing variable.

Loss aversion is a constant. It operates at 11 AM and at 11 PM. But the decision to exit a position does not happen evenly across the day. It clusters around specific conditions - late hours, accumulated fatigue, the isolation of a quiet room with no other person present to slow the thought down.

At 2 PM that same day, Nishant had looked at the same red number and thought: temporary, I'll hold. That thought arrived with moderate force, and he closed the app.

At 11:47 PM, the number was identical. The thought that arrived was different.

The Conditions That Change Decisions

Sleep deprivation had compressed his time horizon without his awareness of it.

A brain running on shortened sleep is measurably worse at evaluating future outcomes. It weighs present discomfort more heavily than future recovery - not because the future seems less likely, but because the present feels more urgent and the future feels like it belongs to a different, calmer version of yourself that does not seem accessible right now.

Decision fatigue had removed the friction that ordinarily protects against wide impulsive choices. By late evening, the cognitive resources that would slow a significant financial decision - sleeping on it, calling someone, reviewing the original investment thesis - were depleted. The path of least resistance was the button two taps away.

Isolation had removed the social check. Investors who make significant decisions with another person present - a partner, a friend, anyone at all - consistently make fewer exits at the wrong moment. Not because the other person has better information. Because presence alone slows the thought down.

A quiet room at midnight removes that check entirely and replaces it with silence that amplifies rather than interrupts.

The market was the same market it had been at 2 PM. The conditions surrounding the decision were completely different.

The Forty Seconds

Inside those forty seconds with his thumb hovering, two thoughts moved back and forth without completing.

What if it falls more? That thought arrived with full texture. With the specific weight of something almost certain. It had a clear action attached - press confirm, lock the loss, stop the bleeding, sleep. The relief was already present in anticipation before anything had been decided.

What if it recovers? That thought arrived softer. Future tense. Under the conditions of that specific hour, the brain systematically discounts future-tense outcomes.

This is not a failure of discipline or a character weakness. It is the stress response doing exactly what it was built to do - prioritise immediate threat resolution over long-term outcome evaluation.

The same wiring has been correct for most of human history. It is poorly calibrated for a brokerage interface at midnight after three weeks of accumulated checking.

The asymmetry between those two thoughts at 11:47 PM was not a function of what the market was likely to do. Both outcomes were roughly equally probable. The asymmetry was a function of the hour, the fatigue, the accumulated cortisol load, and the quiet room with no one else in it.

Nishant did not press confirm.

Not because the fear lifted. It did not. The urgency broke slightly - the way held breath eventually has to release - and in that small gap, he put the phone face-down on the bed without completing the action. He did not sleep well. He checked at 7 AM. The number was the same.

What he had avoided was not necessarily the wrong financial decision. The position may yet fall further.

What he avoided was making a permanent decision from inside a temporary condition - one that would still be in effect in three years, made by a version of himself running on three weeks of accumulated stress and insufficient sleep in a room that was quiet in the worst possible way.

The Pattern

This pattern is not specific to Nishant. It repeats across every significant market correction with a near-identical structure.

Retail investors hold through the early weeks of a drawdown when the stress load is still manageable, and the drop still feels temporary. Across sustained red, the accumulation builds.

The investor who built a perfectly sound strategy in a calm moment - who understood loss aversion intellectually, who had read about staying invested through corrections, who could explain the principle to someone else that morning - opens the sell screen at the point of maximum personal stress.

Which reliably arrives near the point of maximum market drawdown.

The exit is not caused by the market being worse than the investor's strategy can withstand. It is caused by the investor being more tired than their stress management can withstand.

The strategy survived the market. The person did not survive the hour.

Every major correction produces the same exit pattern: retail money leaves equity positions heaviest not at the start of the decline, when the rational case for exiting would be strongest, but weeks in, when the accumulated load peaks and the conditions for a midnight decision are fully assembled.

Markets recover. Investors who exited near the bottom lock in permanent losses on positions that would have returned.

The loss is not caused by a bad strategy or insufficient knowledge. It is caused by the specific combination of fatigue, isolation, and late-hour decision-making that nobody accounts for when they build a long-term investment plan.

The Adjustment

Nishant still holds the fund.

After that November, he made one change. Not to his strategy, not to his asset allocation, not to which fund he was in. He stopped checking the portfolio after 9 PM.

Not as a rule, he disciplines himself to follow. As something he understood specifically about himself after those forty seconds: the version of him that exists at midnight with a red screen is not equipped to make a decision that will still be in effect in 2027.

That version is running on three weeks of accumulated stress, shortened sleep, and the specific silence of a room where no one else is watching.

That version is trying to accomplish one thing only.

Make the feeling stop.

The Real Risk

Permanent decisions made to escape temporary conditions are the most expensive ones in a long-term investment portfolio.

Not because the decision is always wrong - sometimes the position genuinely should be exited. But the conditions that produced the decision at midnight are not the conditions under which the original investment thesis was evaluated.

The thesis was built in daylight, with a clear time horizon, by a version of the investor who had slept.

The market will dip again. Every investor reading this will face a version of that Tuesday in November.

The question is not whether your strategy is sound. Most retail investors who panic-sell have a sound strategy.

The question is what time it is when the dip arrives, whether you have slept, how many weeks you have been carrying the accumulation, and whether the room is quiet in that specific way.

Set the cutoff hour
Write your exit conditions before you enter the position
Make the significant decision in daylight with another person present

The market does not break most long-term investors.

The hour does.

Regards
Chandravanshi Inc
Mr Chandravanshi & Nishant Chandravanshi are the pen names of Nishant Kuamr.
Mrs Chandravanshi & Deepa Chandravanshi are the pen names of Kuamri Deepa Raj.

The Hidden Cost of Leaving Home for Success

Mr Chandravanshi — Sun, 29 Mar 2026 04:41:19 +0000

He didn’t leave Bihar for a better life.

He left because staying started feeling like falling behind.

That decision made sense in 2018.

It still makes sense for most people today.

New city. First job. Shared room. Then a rented one.

Long workdays. Quiet evenings.

Progress became visible.

Salary went up. Independence followed.

From the outside, this is exactly how growth is supposed to look.

But something else starts shifting—and it doesn’t show up on paper.

Calls from home get shorter.
Not because love has reduced.

Because time got structured.

Earlier, conversation was part of life.

Now it has to be scheduled.

And anything that isn’t scheduled… slowly disappears.

This is where most people misread their own situation.

They assume the connection is stable by default.

It’s not.

It’s maintained through frequency, not intention.

Daily presence becomes occasional calls
Occasional calls become updates
Updates become silent with reassurance

“Sab theek hai.”

Two words. No depth.

The system you step into rewards one thing very clearly—output.

Hours worked. Targets met. Income increased.

And slowly, you start optimising your life around that.

Even your relationships.

You send money home.

You think—“Main ye sab unke liye hi kar raha hoon.”

Which is true.

But incomplete.

Because financial contribution is measurable.

Emotional presence isn’t.

So one starts replacing the other, without you noticing.

This is not a personal failure.

It’s a structural shift.

In 2005, proximity defined relationships.

In 2025, coordination does.

And coordination always loses to urgency.

Work has deadlines.

Family doesn’t.

So work wins—every single time.

The cost doesn’t arrive immediately.

It accumulates.

In missed calls, you planned to return.

At festivals, you experience through photos.

In visits where you feel slightly out of place.

Not because they changed.

Because you weren’t there when they did.

Then one moment reframes everything.

You go home.

Same house. Same walls.

But the timeline has moved.

Parents look older than you remember.

Conversations feel shorter than they used to be.

And someone says, casually—

“Ab wo pehle jaise nahi rahe.”

No event triggered it.

No single decision caused it.

Just time… passing without you in it.

This is the hidden trade-off most professionals don’t model.

Growth requires distance.

But distance reduces unplanned interaction.

And unplanned interaction is where relationships actually live.

So the equation looks like this:

More success → less presence → weaker everyday connection

Not broken.

Just thinner.

The uncomfortable part?

You don’t feel it when it’s happening.

You feel it when you pause.

The real question isn’t whether leaving was right.

For many, it’s necessary.

The real question is whether you are actively maintaining what distance naturally erodes.

Because distance doesn’t end relationships.

It removes default access to them.

And anything that requires effort—

competes with everything else that already demands it.

Most people don’t lose connection in one moment.

They slowly stop showing up in the small ones.

By the time they notice—

There’s nothing dramatic to fix.

Just a quieter version of something that used to be fuller.

Deploying Your Salary, Debugging Your Life

Mr Chandravanshi — Sat, 28 Mar 2026 17:43:07 +0000

Why does every "upgrade" still leave the same old constraints running?

The First Signal

The salary hits the account. For about ninety seconds, the number feels real.

Then the outflows run: rent, EMI, school fees, insurance, broadband. By the time the last one clears, sixty percent of the month is already committed. The remaining forty has to cover everything else.

This happens at 60,000 a month. It happens at 2,00,000 a month. The percentage changes slightly. The feeling does not.

Most developers reading this will recognise the sequence. Most will also have attributed it to personal spending decisions. That attribution is the thing worth debugging.

The article is written by Mr Chandravanshi

The Core Problem

The loop is not a spending problem. It is a coupling problem.

Income and obligation are coupled variables. When one grows, the other adjusts to match.

The mechanism runs like this. Income increases. The lifestyle upgrades to reflect the new position - a better flat, a better phone, a school with better infrastructure. Each upgrade is individually logical. Each converts the income increase into a fixed monthly commitment before the increase can widen any gap. The result: the distance between what comes in and what must go out stays approximately constant regardless of the income level.

This is not a discipline failure. Discipline operates on choices. Most of what expands when income grows is not discretionary spending - it is aspirational infrastructure, the category of upgrades that feel like the natural next step rather than indulgence.

A 40% salary increase absorbed into a better flat, a car upgrade, and a private school fee is not recklessness. It is the logical deployment of the increase. The loop continues at a higher absolute level. The margin stays the same.

The Structural Layer

The EMI Layer Makes This Structural, Not Cyclical

A spending problem can be corrected in the next cycle. A structural problem runs underneath every cycle.

EMI culture converts future income into present-tense commitments at the moment of signing. A home loan at 35 does not just shape this month. It constrains career decisions, risk appetite, and investment behavior until 55. Every salary spike across those twenty years - every appraisal, every bonus, every stock vest - gets measured first against the committed outflows before anything else is considered.

The loop does not run month to month. It runs across decades. And every income growth event that should widen the gap instead triggers an upgrade cycle that re-tightens it.

This is the threshold effect. Below a certain obligation density, income growth actually widens the margin. Above it - which most urban professionals cross somewhere around their second EMI - income growth gets absorbed before it can compound.

Most people cross that threshold without noticing. The upgrade that put them above it felt like a reasonable next step.

Why More Income Does Not Help

Why Earning More Does Not Close The Loop

The standard mental model: earn more, obligations stay fixed, gap widens, breathing room increases.

The actual sequence: earn more, lifestyle calibrates to the new position, new obligations expand to consume the increase, gap stays constant.

The model fails because it treats obligations as fixed. They are not. They are a function of the income level and the social position that income level signals. In urban India, lifestyle is legible - the area you live in, the school your child attends, the vehicle you drive are read as information by employers, colleagues, and family. Stepping back from a position when income could support moving forward is visible and carries social cost.

So the loop has a social enforcement layer underneath the financial one. Even when the financial logic of restraint is clear, the social logic of the upgrade is louder.

This is why people who earn twice what they earned five years ago report roughly the same subjective financial pressure. The numbers doubled. The architecture did not change.

The Constraint Of Reality

The Honest Limit Of This Diagnosis

The loop can be interrupted. Not by income growth - the evidence against that is the subjective experience of every developer who has tried.

It gets interrupted on the obligation side. Deliberately not upgrading lifestyle at the pace income growth makes possible. Treating bonus income as structural repair - accelerated loan repayment, obligation reduction - rather than upgrade signal. Holding the flat longer than the social logic suggests. Choosing the school that is good enough rather than the school that positions correctly.

Each of these has a social cost that is harder to calculate than the EMI but just as real.

The architecture was not designed to make these choices easy. The default path - the reasonable path, the socially legible path - is the one that keeps the loop running. Most people take it. Not because they are undisciplined. Because the system was built so the default and the trap are the same thing.

The Conclusion

Financial tightness at a good income level is not a personal configuration error.

It is the loop running correctly. Income and obligation are coupled. Upgrades are logical. EMIs pre-commit future income. Social enforcement makes restraint costly. The gap stays narrow by design.

The fix is not a higher salary. It is a lower obligation density - and that requires making choices the system is designed to make expensive.

Debug the architecture, not the person running on it.

Why India’s Tech Ecosystem Isn’t Converting Hype Into Deep Innovation

Mr Chandravanshi — Sat, 21 Feb 2026 19:41:27 +0000

Why India’s Tech Ecosystem Isn’t Converting Hype Into Deep Innovation

India has engineers.
It has startups.
It has a capital.
It has headlines.
What it doesn’t consistently have is innovation depth.
That distinction matters more than the hype cycle suggests.

The Energy Is Real — But So Is the Gap

India’s technology ecosystem is not small or stagnant.

It includes:
• One of the world’s largest developer populations
• A globally integrated IT services industry
• A rapidly expanding startup base
• Active venture funding
• Strong representation in global tech leadership

On the surface, this should translate into sustained deep-tech breakthroughs.

Yet foundational innovations — in advanced semiconductors, core AI model architectures, operating systems, frontier hardware — still tend to originate elsewhere.

The gap is not active.
It is structural depth.

Services DNA Still Shapes Incentives

India’s modern tech expansion began with outsourcing and IT services.

That created:
• Delivery discipline
• Cost efficiency
• Execution reliability
• Global client integration

Those strengths built scale and credibility.
But they also shaped the ecosystem’s instincts.

Services reward:
• Predictability
• Speed
• Margin control
• Risk minimisation

Deep innovation rewards the opposite:
• Long R&D timelines
• Tolerance for failure
• High upfront burn
• Uncertain outcomes

Ecosystems evolve around what they reward.
For decades, execution was rewarded more than exploration.

Capital Chases Velocity, Not Uncertainty

Venture funding in India has expanded significantly.

But much of it gravitates toward:
• Consumer growth
• Market capture
• Quick scaling
• Valuation acceleration

Deep tech — whether in semiconductors, advanced materials, foundational AI models, or hardware design — requires patient capital.

Patient capital waits through:
• Multi-year research cycles
• Low initial revenue
• High technical risk
• Delayed liquidity

That type of funding is still limited relative to growth capital.

Speed dominates.
Depth requires patience.

Infrastructure Density Is Uneven

Breakthrough innovation does not emerge in isolation.

It requires:
• Advanced research labs
• High-performance compute clusters
• Semiconductor fabrication capabilities
• Tight university-industry integration
• Strong IP enforcement frameworks

India has pockets of excellence — world-class institutes and research centres.
But innovation ecosystems compound when infrastructure is dense and interconnected.
Fragmentation slows compounding.
A single lab cannot substitute for a networked research environment.

Talent Exists — Environment Determines Output

India produces exceptional engineers.
Many lead teams globally.

Yet a significant portion of frontier research still occurs in ecosystems with:
• Higher R&D budgets
• Mature deep-tech funding cycles
• Strong institutional collaboration
• Regulatory clarity

Talent moves toward infrastructure.
Infrastructure amplifies talent.
When the environment supports deep research, outcomes change.
When it prioritises short-term execution, so do outputs.

Education Scale ≠ Research Culture

Engineering graduation numbers are high.

But deep innovation depends on:
• Strong doctoral pipelines
• Commercialisation pathways for university research
• IP translation systems
• Risk-tolerant academic funding

Training developers is not the same as building researchers.
Coding skills build applications.
Research culture builds foundations.
That difference shapes long-term innovation capacity.

Regulatory Friction Slows Experimentation

Emerging technologies thrive on clarity.
Ambiguity increases perceived risk.
Slow procurement cycles delay validation.
Unclear compliance frameworks discourage frontier experimentation.
Innovation doesn’t require deregulation.
It requires predictability.
When rules are unclear, capital retreats to safer categories.

Market Size Isn’t the Same as Market Sophistication
India’s consumer base is enormous.
But deep innovation often depends on:
• High-spending enterprise customers
• Early adopters willing to test experimental technology
• Sophisticated industrial buyers
When purchasing power varies widely, startups optimise for:
• Cost efficiency
• Rapid scale
• Mass-market fit
Not a breakthrough performance.
Market incentives shape product ambition.

The Metrics Distortion
Headlines often highlight:
• Funding rounds
• Valuation growth
• User acquisition
But structural depth is measured by:
• Core IP ownership
• Patent commercialisation
• Advanced infrastructure build-out
• Deep-tech exports
• Long-horizon R&D investment
What ecosystems measure, they prioritise.
When valuation becomes the proxy for innovation, surface growth outpaces foundational research.

This Isn’t a Talent Problem
It’s tempting to frame the issue as a skills gap.
That explanation is simple — and inaccurate.
The constraint is systemic:
• Incentive alignment
• Capital patience
• Infrastructure density
• Research integration
• Institutional coordination
Individuals operate within systems.
Systems determine aggregate output.

What Would Actually Shift the Trajectory?
Structural change would require:
• Expanded long-term R&D capital
• Stronger university-industry commercialisation pipelines
• Scaled compute infrastructure
• Coordinated semiconductor strategy
• Policy stability for frontier sectors
• Incentives for IP ownership, not just user growth
These are infrastructure moves.
Not motivational campaigns.
Not branding exercises.
Not narrative shifts.
**
Final Thought**
India’s tech ecosystem is energetic, ambitious, and increasingly global.
But hype compounds faster than infrastructure.
Visibility scales faster than research depth.
Execution strength has carried the ecosystem far.
To produce sustained deep innovation, the next phase must prioritise:
• Patient capital
• Dense infrastructure
• Research culture
• System alignment
Innovation isn’t only about smart people.
It’s about durable systems that allow intelligence to compound.
And systems — not headlines — determine whether hype turns into history.

Why AI Infrastructure Gaps Are a Bigger Problem Than Talent Shortages

Mr Chandravanshi — Sat, 21 Feb 2026 19:34:30 +0000

Why AI Infrastructure Gaps Are a Bigger Problem Than Talent Shortages

Everyone talks about a lack of AI talent.
“We don’t have enough machine-learning engineers.”
That’s a comfortable narrative — and it’s misleading.
The real bottleneck isn’t engineers.
It’s infrastructure.
And infrastructure problems don’t crash dramatically. They stall quietly.

The Talent Story Is Easy — and Shallow

Blaming talent feels productive:
• Train more engineers
• Launch more AI courses
• Recruit experts globally

It sounds achievable, like a checklist.
But here’s the issue:
Countries and companies with excellent AI education pipelines still struggle to deliver AI at scale.
Why?
Because AI doesn’t run on talent alone — it runs on systems.

AI Depends on Hidden Systems

A successful AI project isn’t just about models and algorithms.

It depends on layers that most people overlook:
• Reliable data pipelines
• Clean and labeled datasets
• Access to modern GPUs and stable compute
• Scalable storage
• Governance and compliance frameworks
• Production deployment environments

If any of these layers fail, the whole system grinds to a halt.
You could hire 100 machine-learning engineers — and none of them would fix broken infrastructure.

Compute Inequality Is Real

AI is exceptionally compute-hungry.

Training and delivering state-of-the-art models requires:
• High-end GPUs or TPUs
• Massive storage systems
• Stable power and cooling
• High-speed networking

Regions without these capabilities aren’t lagging because they lack talent.
They lag because they lack compute density — the hardware, energy, and networking that modern AI demands.
Infrastructure doesn’t just enable capability — it compounds advantage, much like capital does in finance.

Enterprise AI Fails Quietly

Inside many organisations, the pattern repeats:

A team builds a promising prototype. The demo works beautifully. Leadership greenlights expansion.

Then reality hits:
• Data silos block ingestion
• Legacy systems block integration
• Security policies slow deployment
• Procurement delays cut weeks into months
• Compliance requirements stall releases

The model itself isn’t the problem.
The system around it is.
This isn’t a talent gap.
It’s a systems gap.

Why Infrastructure Lags

If infrastructure is the real bottleneck, why don’t companies fix it first?

Because the incentives don’t reward it:
• Hiring AI talent looks progressive
• Investing in data pipelines looks boring
• Building governance frameworks feels slow
• Infrastructure doesn’t show results this quarter

Short-term optics often beat long-term capacity building.
So organisations stack teams without foundations.
And the imbalance surfaces later — when prototypes fail to scale.

Emerging Markets Face Structural Constraints

In emerging ecosystems, the problem deepens.

Even with skilled engineers:
• GPUs are scarce or expensive
• Cloud compute costs are high relative to local budgets
• Regulatory ambiguity slows planning
• Digital ecosystems are fragmented

The environment itself restricts scale.
Talent leaves for better infrastructure.
Infrastructure stagnates.
That’s why people say, “We are behind in AI.”
It’s not education that’s missing — it’s infrastructure depth.

The Illusion of AI Readiness

Many organisations believe they’re AI-ready:
• They have data scientists
• They run pilot models
• They publish roadmaps

But readiness isn’t about headcount.

True AI readiness means:
• Unified and clean data
• Scalable compute environments
• DevOps-integrated deployment pipelines
• Continuous monitoring
• Clear governance and security

Without these, AI stays experimental, not operational.

Infrastructure Is Slow. Talent Is Visible.

Training engineers takes months.
Building infrastructure takes years.

Infrastructure demands:
• Capital investment
• Cross-department coordination
• Long-term planning
• Institutional alignment

That’s why it’s politically and operationally harder.
So companies see talent.
Infrastructure stays invisible.

What Builders Should Notice

If you’re building AI systems, pay attention:

Ask yourself:
• Are failures in modeling or deployment?
• Are issues algorithmic or architectural?

Most AI friction isn’t about accuracy.

It’s about:
• Data flow
• Environmental stability
• Organisational bottlenecks
The real engineering challenge isn’t smarter models.
It’s a system that actually runs them.

Why This Matters in the Long Run

Infrastructure compounds advantages:
• Strong systems amplify talent
• Innovation cycles shorten
• Deployment becomes routine
• Organizations scale confidently

Without infrastructure:
• Talent works in isolation
• Prototypes remain demos
• Scaling becomes prohibitively expensive

Infrastructure determines velocity.
Talent determines direction.
Without a foundation, direction doesn’t matter.

Final Thought

The global AI race is often framed as a competition for talent.
That’s only part of the story.

The deeper competition is over:
• Compute capacity
• Data architecture
• Institutional alignment
• Long-term infrastructure investment

Talent matters — but it cannot overcome systemic bottlenecks by itself.
If you want a durable AI capability, you build the foundation first.
Models will improve every year.
Infrastructure decisions last a decade — and systems always outlast individuals.

Regards,
Nishant Chandravanshi