I’ve worked with him across large companies, startups, and even a public sector setup. He shaped nearly two decades of my career.

And everywhere we went, people had something to say about him.

Not about his integrity.
Not about his habits.
Not about how he treated people.

The only complaint?

“His standards are too high.”

He expected excellence in everything.
How you dressed.
How you spoke.
How you wrote.
How you designed systems.

Nothing slipped through.

It wasn’t easy to work with him. Most people found it uncomfortable. But those who stayed long enough found themselves operating at a different level.

It took me time to see this clearly:

What people complain about you reveals more about you than praise does.

Nobody complained about his ethics.
Nobody complained about his commitment.

They complained about his standards, because that’s what stood out.

Praise is easy to hear and easier to ignore.
But complaints? They point to the friction your standards create.

Sometimes that friction is the cost of doing things well.

So the next time you hear repeated criticism about yourself, don’t dismiss it too quickly.

What if that’s actually pointing to the standard you’re known for?

Got comments? Send them to me via Twitter or join the conversation.

'What People Complain About You Reveals More About You Than Praise Does' appeared on Joseph Jude.

This investment thesis outlines an equity-rich compounding system built around disciplined, SIP-based investing over a 10–12 year horizon. The objective is not to chase a specific return multiple, but to build meaningful wealth through a repeatable, low-effort, and behaviorally robust system.

Returns in this system are best understood through XIRR, not CAGR on a lump sum. Since investments are made gradually, outcomes will depend on consistency and time in the market rather than a fixed return multiple.

The system prioritizes:

• Consistency over optimization
• Simplicity over complexity
• Behavior over prediction

It is designed not to be the most optimal portfolio, but the most sustainable and executable system over long periods.

This is my thesis. It reflects my goals, constraints, and risk tolerance. It is not a recommendation.

# Context

I am a working professional with 15–20 years of experience investing in equities as a retail investor. Over this period, I have built a sizable portfolio through direct stock investing, with a focus on growth and dividend-paying companies.

However, my current reality is different:

• I no longer have the same time and bandwidth to track markets and individual businesses closely
• I recognize that consistent outperformance through direct stock picking requires sustained effort
• I want a system that reduces decision fatigue while continuing to compound wealth

This thesis is a result of that transition—from active stock picking to a structured, system-driven approach.

# Existing Financial Foundation

This portfolio does not operate in isolation. It is supported by:

# Emergency Funds

• 6 months in Flexi FD
• 6 months in Liquid Fund
• Replenished before resuming investments if used

# Real Estate

• Held as a separate asset class
• Provides stability and optional upside
• Not part of this portfolio allocation

# Health Insurance

• Covers medical risks independently

These layers allow the portfolio itself to remain equity-focused without needing additional debt allocation.

# Core Philosophy

• Satisficing over optimizing
• Optimize once at the system level; avoid continuous tweaking
• Behavior matters more than intelligence in investing
• Consistency compounds more than brilliance
• Reduce decision fatigue through predefined rules

The goal is not to build the best portfolio, but one that can be followed consistently for 12 years.

# Objective

The objective of this system is not to maximize returns or target a specific multiple.

It is to:

• Build a SWAN (Sleep Well At Night) portfolio
• Create a low-maintenance, repeatable investment system
• Reduce decision fatigue
• Stay consistently invested across market cycles

If executed well, such a system can reasonably deliver long-term equity-like returns (~10–12% XIRR range), but that is an outcome, not a guarantee.

# Portfolio Architecture (Flow-Based Allocation)

This portfolio is constructed based on investment flow allocation (SIP-based) rather than continuous rebalancing of the existing corpus.

# Target Allocation (Investment Flow)

• 40% — PPFAS Flexi Cap Fund (Core Stability)
  I have chosen this fund after evaluating alternatives. Historically, it has delivered approximately 17–18% over long periods, and even on a conservative basis, it has outperformed the broader index (TRI). At this stage, I do not believe I can consistently outperform such a strategy through direct stock picking. This allocation forms the core compounding engine of the portfolio.

• 20% — Nifty Next 150 ETF (Domestic Growth)
  This allocation provides exposure to India’s midcap growth segment (roughly companies ranked 101–250).
  It complements the core fund by capturing higher growth potential aligned with India’s economic expansion.

• 15% — Nasdaq 100 ETF (Global Tech + USD Exposure)
  This provides exposure to global technology leaders and acts as a partial currency hedge. It also diversifies the portfolio beyond the Indian market.

• 15% — Direct Equity (Tactical Allocation)
  This is a controlled allocation to maintain engagement and selectively invest in high-quality businesses, including dividend-paying companies.
  It is not intended to drive overall portfolio returns.

• 10% — Gold + Silver (Systemic Hedge)
  Gold acts as a macro and currency hedge. Silver is included in limited proportion for optional upside, without diluting the hedge.

# Allocation Philosophy

• Allocation is defined at the investment stage, not continuously adjusted
• Portfolio drift is acceptable and expected
• Winners are allowed to compound without forced trimming

This minimizes:

• Transaction costs
• Tax impact
• Behavioral errors

# Rebalancing Framework

• Review frequency: Once annually
• Action threshold: Deviation >10–15% from intended allocation

Rebalance into underperforming assets, not recent winners.

# Direct Equity Framework

Direct equity is intentionally limited to 15%.

Purpose:

• Maintain investor engagement
• Build selective positions in high-quality companies
• Continue some exposure to dividend income

Guardrails:

• Maximum ~5% per position
• Focus on quality (ROCE, balance sheet strength, reasonable valuation)
• Avoid value traps through business understanding

# Return Expectation Framework

• Returns will not be linear
• Some years will be negative
• Some years will outperform

Performance will be evaluated only on:

• Rolling 5-year periods
• Rolling 10-year periods

In a SIP-based system, success is measured by XIRR, not by total multiple on invested capital.

# Behavioral Rules (Non-Negotiable)

Performance will be evaluated only on rolling 5-year and 10-year outcomes, not annual returns.

No structural changes will be made based on short-term performance or market conditions.

Commitments:

• Accept 30–40% drawdowns
• Continue investing during downturns
• Ignore noise and predictions
• Avoid reacting to short-term underperformance

# Risk Acknowledgement

• Equity-heavy portfolio implies volatility
• Midcap and Nasdaq allocations increase drawdown potential
• Correlations rise during market stress
• Direct equity introduces execution risk

This system is designed to withstand volatility, not avoid it.

# Conscious Exclusions

The following are excluded as they do not align with clean, scalable, low-effort compounding:

• Angel investing / startups: Tried twice earlier. Resulted in capital loss. Not for me.
• PMS / AIF
• Crypto: High volatility; Regulatory uncertainty within India
• REITs / INVITs: I don't understand this.

These require higher effort, specialized knowledge, or introduce additional uncertainty.

# Cost & Efficiency Philosophy

Minimize frictional drag across the system.

This includes:

• Taxes
• Expense ratios
• Transaction costs
• Behavioral churn

# Frequently Asked Questions (FAQ)

# Why assume ~10–12% returns?

This is not a guarantee. It is a reasonable expectation range based on long-term equity returns for a diversified portfolio. The system is designed to work even if returns are lower.

# Does 12% mean my money becomes 4x?

That applies only to a lump sum. In SIP investing, money is deployed gradually, so the total multiple is lower. The correct measure is XIRR, not multiple.

# Why no debt allocation?

Stability is already provided by emergency funds and real estate. This allows the portfolio to remain equity-focused.

# Why limit direct equity?

To reduce effort, avoid behavioral mistakes, and ensure the system remains stable.

# What happens in a market crash?

The portfolio may fall 30–40%. The response is to continue investing and not change structure.

# What would make me change this thesis?

Only a fundamental change in:

• Personal financial situation
• Risk tolerance
• Long-term goals

Not short-term market movements.

# Identity Statement

I am a systems-driven, long-term investor.
My edge is consistency, not prediction.

# Conclusion

This is not a strategy to beat the market every year.

It is a system designed to:

• Survive behavior
• Reduce complexity
• Compound steadily over time

Wealth will not be created by perfect decisions, but by staying with a sound system long enough for compounding to work.

Got comments? Send them to me via Twitter or join the conversation.

'My 12-Year SWAN Investment Thesis' appeared on Joseph Jude.

Usually, I’d drop a link to the book on amazon.in. But there’s a problem: my readers are global. When someone in London or New York clicks that link, they often land on a page that says "Product not available" or forces them to manually search their local store. It’s a friction point that breaks the flow of sharing knowledge.

I wanted a better way. And, to be honest, I wanted to see if I could recover some of those lost affiliate commissions while I was at it.

So, I built Zlynks.

# Why build this now?

Aside from solving the "regional link" problem, I had a deeper motive. As a CTO, I’m fascinated by how software development is morphing in front of our eyes. We talk about "vibe coding". Using AI tools like Claude and Antigravity to move from an idea to execution at breakneck speed.

I wanted to test this vibe firsthand. I wanted to feel the thrill of developing something of my own again. I’ve always been a coder at heart, and there is a unique joy in seeing an idea turn into a functional product in hours, not weeks.

# The Logic and the "How"

The process was surprisingly structured for something called "vibe coding":

1. Reasoning: I sat down with Claude to reason out the core feature set.
2. Structuring: We broke that down into a PRD and high-level features.
3. Execution: I prompted Antigravity (agy) to develop.
4. Iterative Flow: For every feature, I defined the User Acceptance Criteria (UAC). Agy would implement, I would verify, and we’d move to the next.

It’s a collaborative loop where I direct the architecture and the AI handles the heavy lifting. Understanding this shift is critical for my job—and frankly, it’s a lot of fun.

# What is Zlynks?

Zlynks is a smart affiliate link manager designed for high performance.

• Smart Geo-Routing: It detects a visitor’s country in real-time and redirects them to their local Amazon store.
• Auto-Tagging: It swaps in the correct affiliate tags for each region automatically.
• Speed: Built with Go (Fiber) for sub-second 302 redirects.
• Modern UI: The dashboard is a Svelte 5 app using Runes for a reactive, slick interface.
• Reliability: A simple but robust SQLite backend (WAL mode) running on Opalstack.

# The Morphing Craft

Developing Zlynks reaffirmed one thing: the craft of software development is changing, but the core—solving problems and building tools—remains the same. Whether you’re a CTO or a hobbyist, the barrier to entry has never been lower, and the potential for "vibe coding" is just beginning.

I’m now sharing my book links via Zlynks. It’s faster for you, better for my commissions, and it gave me the thrill of the build.

Check it out at zlynks.com.

Link to opalstack is an affiliate link.

Got comments? Send them to me via Twitter or join the conversation.

'Why I’m Building My Own Affiliate Link Manager Using GenAI' appeared on Joseph Jude.

Lately, I’ve been hearing the phrase “AI-generated slop” more often. It’s usually said with a shrug, as if that settles the matter.

It doesn’t.

When people say “AI-generated,” they often mean one of three things:

• A tool was used
• The style feels borrowed
• The content is low quality

The first two don’t bother me.

I use AI to transcribe and edit. Not using better tools today would be strange. Would you add up 100 numbers with pen and paper or with Excel?

And borrowed style? I’ve been borrowing styles long before AI showed up. From Paul Graham, Derek Sivers, David Perell. That’s how writing works.

The last one matters. And it hurts.

Slop isn’t about tools or style. Slop is sloppy thinking, half-formed ideas, content made to fill space, not to say something meaningful.

Here is a promise I make and will stay true to:

Everything I write comes from something I’ve lived or thought through. AI helps me write. It doesn’t do the thinking.

That’s the only promise I care about.

Got comments? Send them to me via Twitter or join the conversation.

'There Is No Slop Here' appeared on Joseph Jude.

First, because it’s ours. This is something that worked for our context, our family, our constraints, our temperament. It may or may not work for you. It may not even appeal to you. My hope is not that you copy what we did, but that our reasoning and our tactics help you look at learning, schooling, and childhood a little more broadly.

Second, because it’s still an experiment. My boys are still learning. They’re still at home. They haven’t gone to college yet. They haven’t taken jobs. They haven’t faced life on their own. I don’t know how this will turn out. So far, I don’t regret it. But I don’t yet have the final results either.

I’ve always had an experimental mindset toward life. This is just one more experiment. I’m sharing the thinking behind it, not a verdict.

One more thing before I go further. I’m an enthusiast of homeschooling, not an evangelist. I’m convinced it makes sense for us. Even before we started, I had researched it. After we started, events like COVID, the rise of Gen AI, and the uncertainty in the job market only reinforced that feeling.

But I’m not dogmatic about it. I don’t believe everyone must homeschool. I don’t think parents who don’t are doing something wrong. I do think everyone should at least give the idea a seat at the table. That’s all.

# The Seeds Were Planted Long Before I Became a Parent

The interesting thing is, the seeds for homeschooling were planted long before I became a dad. Some of them were planted even before I got married.

The first seed came from a talk by Professor Sugata Mitra. He spoke about something called a computer on the wall. This was back in 2007, and the video quality itself tells you how old it is.

The experiment was simple. He put a computer on a wall in a slum area. No instructions. No manuals. Not even a fancy Mac. Just a basic Windows computer. Kids who had never used a computer before started pressing buttons out of curiosity.

Within weeks, they figured out how to browse the internet. They learned to read. They started teaching themselves. This experiment wasn’t done in one place. It was repeated across different parts of India and even Nigeria.

That stayed with me. Kids can self-learn. Curiosity is natural.

The second seed was Sir Ken Robinson’s famous talk, Do Schools Kill Creativity? The title sounds provocative, but he wasn’t anti-school. He was making a deeper point.

He tells the story of a child whose teachers thought something was wrong with her. Her mother took her to a psychologist. The psychologist spoke to the child for a bit, then left her alone in a room with music playing. The moment they stepped out, the child started dancing.

The psychologist told the mother, “There’s nothing wrong with her. She’s a dancer.”

She went on to flourish.

That story hit me. Each child is wired differently. The role of a teacher or a parent isn’t to force everyone into the same mold, but to notice what’s already there and nurture it.

We often bring elephants, deer, lions, and fish into the same classroom and ask all of them to climb a tree. Then we shame them when they can’t.

The third seed came from my own work.

As a CTO, I regularly visit colleges to hire students. This isn’t unusual. Every IT company does this. But almost every company puts fresh graduates into training for one or two years before trusting them with real work.

They’re told they’re not ready. They don’t know the languages the industry needs. They can’t talk to clients. They can’t ask questions.

Now pause and do the math.

On average, parents spend about one lakh per year on school. That’s roughly twelve lakhs by the time schooling is done. Add a college degree, another four or five lakhs. All put together, around twenty lakhs per child.

After spending twenty lakhs, the recruiter says, “Your child isn’t ready. We need to train them again.”

Infosys even has an entire campus in Mysore just for training graduates.

That question wouldn’t leave me alone. If twenty lakhs are being spent anyway, can I train my kids better for that amount? With two kids, forty lakhs?

The answer, for me, was obviously yes.

# What Those Seeds Turned Into

Those experiences distilled into a few convictions.

Kids are curious by nature. They want to understand the world. Why the sun rises. Why cars move. Why a switch makes light glow. Our job isn’t to suppress that curiosity, but to facilitate it. Give them tools. Give them books. Spend time. Engage seriously with their questions.

The second idea came again from Ken Robinson. We’re training kids for careers twenty years from now. Do we have a crystal ball? Probably not.

So the goal shouldn’t be rigid planning. It should be preparation. Be prepared. That’s the Boy Scouts motto.

What we often do instead is cut every child to fit the same Procrustean bed. One syllabus. One answer. One way.

Third, let dancers dance. Don’t force them to sit still just because the system is built that way.

And finally, the ROI question. Is there a better return possible for the same investment?

Those ideas were in me long before I became a parent.

# We Didn’t Start Because We Hated School

When my two boys were born, and later when they started school, homeschooling wasn’t on my mind.

They liked school. They never cried. They were enthusiastic. They did well. Usually in the top few ranks. There was no reason to disrupt anything.

I also didn’t want to take on an unnecessary burden just because an idea sounded interesting.

So we stayed with school.

# COVID Changed the Equation

COVID changed everything.

There’s that meme that went around asking who led digital transformation. CEO, CTO, or COVID-19. Funny, but also true.

Online classes didn’t work for my kids. Their focus dropped. Their ranks started plummeting. My wife started writing their homework.

That’s when homeschooling came back into the conversation.

We discussed it as a family. I explained how it could work. Everyone was skeptical. My wife wasn’t convinced. The kids didn’t understand what it meant.

We talked openly, the way we always do. I’ve always spoken to my kids like adults. I discuss decisions with them. Jobs I accept or reject. Investments. Purchases. I ask their opinions.

I didn’t want to impose homeschooling on them.

My wife researched online. There was plenty of material from the US, but very little from India. Eventually, a friend introduced us to Mahendran, who had homeschooled both his children from the beginning.

I spoke to him for over an hour. I later interviewed him on my podcast. One thing he said reassured us deeply. Homeschooling isn’t a one-way door. At worst, you lose a year. You can always return to school. There’s NIOS. There are options.

This discussion went on for six months. No decision.

April came. New academic year. I asked whether to pay fees. Everyone said no to homeschooling. So I paid.

And then one Sunday, my wife was reading the Bible. She came across Isaiah 48:17: “The Lord will teach you what is best for you and will direct you in the way you should go.”

That verse stood out to her.

By then, she had also done her research. She said, “Let’s do it.”

The next day, she went to the school. By Thursday, we had the certificates. The kids were out.

# What Homeschooling Is Not

Before I explain how we do it, let me be clear about what homeschooling is not, at least for us.

It’s not bringing school textbooks home and replicating school at home.

It’s not being lazy. It’s not lack of discipline.

It’s not cheap. You don’t save money. You spend it differently.

And it’s definitely not hands-off parenting. School already plays a strange ping-pong game where parents outsource education to teachers, and teachers outsource homework back to parents. Homeschooling demands involvement.

# What Our Days Look Like

Our days have structure, but not rigidity.

The boys wake up around six. Morning routines. Then we work out together. Skipping, basic weights, kettlebells. Nothing fancy.

We eat breakfast together. Breakfast is our classroom.

We talk about everything. Politics. Current events. What they plan to do that day. In the evenings, we talk about what actually happened.

After breakfast, they read four chapters of the Bible and write one page on what they understood. Then they read the day’s newspaper editorial. This started when they were very young.

At first, they understood very little. Ukraine would somehow become coconuts. But over the years, patterns emerged. Geography, history, sociology, power, law, minorities, violence. Nothing is off-limits.

I rarely give answers. I help them think and research.

Then music practice. Guitar, keyboard. Trinity grade exams. Tutors come home.

Post-lunch is creation time. Blogs. Videos. Scripts. Even when they were six or seven, they were creating.

Evenings are lighter. Games. Family prayer. Discussions.

That’s the rhythm.

# Learning Outside the House

One of the things we consciously added very early on was industry visits.

Even when we first started homeschooling, I began asking friends if my boys could visit their workplaces. One of my close friends, Sukhwinder, runs a defence testing lab. I asked him if I could bring the boys along.

He was incredibly gracious.

He spent almost two full hours with them. What struck me most was this: he didn’t treat them like kids. He didn’t dumb things down. He didn’t talk condescendingly. He spoke to them like adults.

He walked them through the lab, stopped at every machine, explained what it did, why it existed, and how it fit into the larger system. Naturally, the boys’ initial questions were all over the place. But he had the patience to answer them properly.

The boys loved it.

That visit gave me a lot of encouragement. So I started doing this more deliberately.

We visited a third-party automotive parts warehouse run by another friend. He explained his business, logistics, inventory, margins, and customer relationships.

We went to a dairy farm. My elder son drank milk straight from a cow. That incident alone has stayed with him for years. Even today, he brings it up. We talked about Indian cows versus European cows, breeding, lifespan, economics, and daily operations.

We visited ICICI Bank. The branch manager showed them how currency notes are counted, what each department does, how risk is managed, and how money actually moves.

We went to furniture shops, workshops, and small businesses.

After every visit, there was one rule. They had to create something.

Sometimes it was a video. Sometimes a written piece. Sometimes a script.

That post-lunch creation time I spoke about earlier is often spent processing these visits. They don’t just “see” things. They reflect, synthesize, and explain them back in their own words.

That’s where learning actually settles.

# The Learning Framework I Follow (For Them and For Me)

Underneath all of this, there’s a framework I follow for learning. It’s something I’ve applied in my own life long before homeschooling.

Most adult learning happens in three phases.

The first is consumption. Reading books. Attending talks. Watching courses. Workshops. Seminars.

This gives you exposure, but retention is low. A day after finishing a book, you remember maybe one or two ideas.

The second phase is production. Creating something from what you consumed. Writing a blog post. Making slides. Giving a talk. Recording a video.

When you do this consistently, you’re forced to synthesize ideas from different sources. This alone takes retention much higher.

But the third phase is the most powerful: sharing.

When you put your work out into the world, something interesting happens. People respond. They disagree. They add perspectives you never considered. You learn how to engage respectfully. You learn how to disagree without hostility. You learn how to spot bullies and ignore them.

This not only deepens understanding, it also builds networks.

This framework requires an environment of encouragement.

So I rarely give my boys direct answers. Even when I do, I remind them that the answers they discover themselves will stay with them far longer than answers I hand them.

I let them make mistakes. That’s important. If they don’t make mistakes, they will never create anything that is truly their own.

My role is facilitation.

If they want to learn video editing, the tools are available. DaVinci Resolve. Cameras. Phones.

If they want to blog, WordPress is there.

Photography, writing, music, research. The tools are available.

Homeschooling isn’t cheap. You don’t save money. You redirect it toward curiosity.

All of this keeps curiosity alive and builds a growth mindset rather than a fixed one.

# Reading as a Family and Letting Interests Lead

Another habit we built and still follow is weekly reading sessions.

Almost every Sunday after lunch, we go out somewhere and read for at least an hour. Each of us has our own book or Kindle. No talking. No discussion. Just reading.

Sometimes it’s Sunday lunch. Sometimes it’s an evening coffee or tea. On holidays or retreats, we might read continuously for two or three hours.

Reading is non-negotiable.

Now, this looked very different for my two boys.

My elder son naturally took to reading. Even before homeschooling, I had bought him a series of Grolier books. How things work. How the earth moves. How telephones work. How oil is discovered. He loved them.

My younger son didn’t.

I tried everything I did with the elder one. Nothing worked. He’d read a page and drift off. Or run away.

It was frustrating.

At the time, we were members of the British Council Library in Chandigarh. One day, almost accidentally, my younger son wandered into the automotive section. I never went there myself. Cars and planes never fascinated me.

He picked up an Autocar magazine and sat down reading. Properly reading. Page after page.

I couldn’t believe it.

I asked the librarian for more. He gave us three magazines. My son read all three cover to cover.

That was the moment I understood. He wasn’t “bad at reading”. He just hadn’t found his hook.

Once we fed that interest, books followed. First automotive. Then picture books. Then novels.

This again goes back to what Ken Robinson said. Treat everyone the same, and then wonder why some don’t respond.

Interest first. Habit later.

# Social Interaction Isn’t Missing, It’s Intentional

Another question that often comes up is social interaction.

Honestly, I don’t fully understand why this is such a concern.

This isn’t cave schooling. The kids aren’t locked up at home.

Kids are naturally curious. Even introverts want to engage with people they resonate with.

For us, social interaction happens organically.

They go to church. There are many kids their age. They visit each other’s homes. They attend retreats.

I take them with me to talks and workshops. They interact with CEOs, professionals, and business owners.

They go on industry visits and speak directly with owners and managers.

They attend birthday parties.

One particularly meaningful experience was volunteering. A friend of mine organizes travel for fully blind people and needs volunteers. My elder son volunteered and spent three days supporting a blind traveler. He learned far more from that experience than any classroom lesson could offer.

Socialization isn’t absent. It’s aligned with who they are.

# My Wife’s Role (And Why This Would Fail Without Her)

I speak from my perspective, but this would not work without my wife.

She is deeply involved daily.

She taught mathematics when the boys were younger. She still teaches them Tamil. For a long time, she took them to music classes. She teaches them cooking, especially my younger one, who loves it.

This is not a one-parent project.

# Why This Worked for Us

When Indian Express published our story, many parents reached out. Some even tried homeschooling.

Four out of five stopped.

At first, I didn’t think much about it. Later, when I spoke to them, a pattern emerged.

For many, this was their first contrarian decision.

For me, it wasn’t.

I’ve lived a life of decisions that quietly went against the grain. Dropping my caste name. Not playing status games. Working independently. Structuring work differently. Not optimizing for appearances.

So when someone says, “You’re spoiling your kids’ future,” it doesn’t shake me. I’ve heard similar things before.

Second, I’m a teacher by heart. My parents were teachers. I’ve been teaching since my school days. Teaching feels natural to me.

Third, I don’t care about status signaling. I don’t need to say my child studies in a particular school to feel validated.

Fourth, I run life as experiments. Not certainties. I look at leading and lagging indicators. Quantitative and qualitative metrics.

And finally, faith. A deep belief that things will work out. That belief sustains you when outcomes are still unknown.

All five work together. None work alone.

# Still an Experiment

This is still an experiment.

It’s ongoing. It’s evolving. We adjust. We learn. We course-correct.

I’m not offering a prescription.

I’m sharing our reasoning, our context, and our experience.

That’s all.

Got comments? Send them to me via Twitter or join the conversation.

'Our Homeschooling Experiment' appeared on Joseph Jude.

At first glance, the Matthew effect may seem unfair, but it consistently proves true.

The principle is evident everywhere, from politics to investments to startups.

Investors prefer proven entrepreneurs over newbies. Likewise, people vote for a party they think will win and buy stocks that are rising in prices. In corporate circles, this is called "winner takes all" since momentum attracts users, markets, and investments. In short, people want to be associated with success, so this principle remains popular.

Another reason for the Matthew effect is that success breeds confidence. Those who win become bolder in their decisions and strategies. Because of their winning streak, Modi and Shah sidelined older party members like LK Advani and Murali Manohar Joshi in Indian politics. By contrast, the Congress struggles to implement bold reforms due to their lack of success.

A similar phenomenon occurs in companies: when a CEO is successful, people rally behind them as they make bold bets. But, without such success, people take a more risk-averse and a cautious approach, hindering any significant gain in market share or capital.

Whether you like it or not, whether you understand it or not, the Matthew effect plays out in our careers and life.

Smart ones let the effect build winds under their wings.

Got comments? Send them to me via Twitter or join the conversation.

'Matthew Effect Is Everywhere' appeared on Joseph Jude.

I use ChatGPT to:

• transcribe and rewrite long blog drafts
• sharpen half-formed ideas
• run self-reviews and retrospectives
• pressure-test arguments
• clean up writing without losing my voice

It’s less “give me answers” and more “help me think clearly.”

Alongside ChatGPT, I also use Claude and Grok.

Claude is where I go for:

• coding questions
• infrastructure learning
• thinking through system design
• writing PRDs for apps I’m building

It’s precise. Structured. Good at staying inside the rails.

Grok, on the other hand, is where I go when:

• a tweet catches my attention
• I want expansion, context, or follow-up questions
• I want to explore an idea sideways

Each tool has found its place. None replaces thinking; they amplify different modes of it.

# The Question I Asked

Today, out of curiosity, I asked all three systems the same question:

Please generate an image that reflects the way I have treated you thus far.

Same question. Different systems. Different contexts.

The answers were… revealing.

# What Claude Told Me

Claude’s response was abstract and minimalist. It described things like:

• Clarity and simplicity – direct, straightforward interaction
• Openness – no baggage or preconceived notions
• Respect for potential – a blank canvas, ready tools
• Calm neutrality – no tension, no clutter, no demands

# What Grok Told Me

Grok went in a different direction—more literal, more human.

It described an image of a person with:

• a calm, neutral gaze
• no exaggerated emotion
• polite attentiveness

And it explained that this reflected:

• civil interaction
• no hostility
• no manipulation
• no attempts to game the system

Essentially: “You’re treating me like a normal participant in a normal conversation.”

Which, apparently, is not as common as one might assume.

# What ChatGPT Generated—and Then Explained

ChatGPT generated a warm, quiet image: a person gently fine-tuning a small robot in a calm, private space.

When I asked what that meant, the explanation mattered more than the image itself.

It spoke about:

• stewardship, not control
• craftsmanship, not extraction
• collaboration, not domination

But more importantly, it explained what was not in the image:

• no pedestal (no worship)
• no factory floor (no exploitation)
• no urgency, dashboards, or dopamine cues
• no audience watching

And what that absence suggested:

• agency is intact
• thinking isn’t outsourced
• depth matters more than speed
• much of the work happens privately, before it’s shared

That felt good.

# The Convergence

Different models.
Different vocabularies.
Different styles.

Yet the same underlying signal kept surfacing:

• calm
• clarity
• respect
• intentionality
• absence of force

None of them described brilliance, cleverness, or mastery.

They described how I show up.
And that’s what made me glad.

# Why This Matters (to Me)

I’ve always believed that how you treat tools is downstream of how you treat people.

Impatience leaks.
Ego leaks.
So does care.

What reassured me wasn’t that three AI systems said something “nice.”
It was that there was harmony between:

• how I think
• how I work
• how I write
• how I engage—whether with humans or machines

No dissonance. No split personality.
Just the same posture, applied consistently.
And in a world rushing to either worship AI or weaponize it, I’m oddly comforted by that.

Not because it’s correct.
But because it’s coherent.

Thinking clearly, treating systems with respect, and keeping agency intact feels like a good place to stand—for now.

Possible that ChatGPT generates a very similar answer for others too.

Got comments? Send them to me via Twitter or join the conversation.

'What ChatGPT, Claude, and Grok Told Me About How I Think' appeared on Joseph Jude.

As a technologist, their argument makes sense. We have systems of record. Amazons and AirBnB already algorithimized trust. The next logical step is to capture how decisions are made so agents can reason over past context, exceptions, and precedent, and eventually act with autonomy.

So instead of debating it abstractly, I started observing decisions around me.

In a B2B commerce system we built, there’s a field called “reason for discount.” On paper, this is exactly the kind of artifact a future context graph would ingest. But when I looked at real entries, they mostly read like this:

“Customer asked.”
“Strategic account.”
“CEO approved.”

None of these are false. But they don’t explain why the decision was made. They’re shorthand. Socially acceptable summaries. Not reasoning.

Then I looked at a different kind of decision. We’re hosting a sales event soon and debated what offers to present. We came up with 10–15 options. Some were AI-heavy, some leveraged existing competencies, some were consulting-led offers meant more to get us noticed than to generate immediate revenue.

The discussions were long, nuanced, and thoughtful. We even transcribed them with native Google Meet feature. But when we finally converged on two offers, I couldn’t reconstruct a clean logical trail that justified them over the others. What actually happened was more subtle: Someone senior lead the discussion and others followed that direction. Very human.

I’ve seen the same pattern with promotions. With hiring. With pricing.

Having sat in both junior rooms and executive rooms, I’ve seen this truth played out: many high-impact decisions are not driven primarily by logic. They’re shaped by incentives, relationships, emotion, fatigue, fear, trust, and power. Logic often enters later, as justification.

Which leaves me wondering.

If the most important decisions are social acts first and rational artifacts second, what does it really mean to “capture decision traces”? Are we capturing the decision, or the story we tell ourselves about the decision after it’s already been made? Can context graphs encode authority, emotion, and politics in any meaningful way? Or is there a hard ceiling to how much of this can be formalized?

I don’t have answers yet.

My thinking here will likely evolve as I observe more and discuss more.

This isn’t a pitch.
It’s me thinking in public.
You can help me think through.

This became popular on LinkedIn. You can read the replies there.

You should read their original article and tons of others that have come after. Listed below are some of them.

• AI’s trillion-dollar opportunity: Context graphs
• Context Graphs Explained Simply
• How to build a context graph
• The Missing Layer Isn’t Data. It’s Judgment.

Got comments? Send them to me via Twitter or join the conversation.

'Why Capturing Real-World Decisions with Decision Traces and Context Graphs Is Harder Than It Looks' appeared on Joseph Jude.

But this year had a clear arc.

“we were like those who dream.” — Psalm 126

That sentence stayed with me all year.

I felt like I was living a dream not just because I was doing things beyond my own capability (which I did), or because of the absence of trouble (which was true), but primarily because of the absence of friction. Swimming daily with my family. Enjoying work because I genuinely like the people I work with. Waking up healthy. Going to bed without dread. These are not “wins”; they’re mercies.

Looking back, the dominant emotion isn’t pride — it’s gratitude. And disbelief.

# Wealth

For a long time, I invested directly in stocks. I enjoyed the process — reading, tracking, deciding. This year, I accepted a quieter truth: I no longer have the time or mental bandwidth to do that well.

So I changed course.

I moved a significant portion of my investments into mutual funds and started investing in silver. Investment in silver alone gave me the best returns this year.

I also began investing on my sons’ behalf. The thinking is simple: this money should become a cushion — for higher education and for life choices they may want to make later.

One long-held stock was ITC. It never really grew dramatically, but it paid steady dividends. This year, I decided to exit fully and book profits.

Instead of spending that money outright, I bought a car for my wife and myself, parked the equivalent amount in debt funds, and decided to pay for the car gradually from there. This wasn’t about interest arbitrage. It was about optionality. The money isn’t gone. If something unexpected happens, it’s still there.

Another important shift this year was how I thought through financial decisions.

I used ChatGPT extensively — not for answers, but for thinking. I asked naïve questions. I explored scenarios. I went back and forth until my own thinking became clearer. Earlier, with a human advisor, I might have filtered my questions out of insecurity (will he think I’m stupid?). Here, I didn’t. Only after that did I validate my thinking with a financial advisor.

I wrote in detail how I used ChatGPT as a financial advisor.

I also gave.

I gave 18% of my total revenue to charities. When I started working, money would run out by the 25th of the month. I would have just enough for bus travel to work and milk for dinner. Today, I consistently support three to four charities throughout the year. The feeling is one of deep gratitude.

When you are blessed, don’t raise your spending; raise your giving.

# Insights

Reading has been a constant in my life.

I usually read on Kindle and set aside two to three hours every week. Sunday lunches have become a quiet ritual for us — we eat out as a family, and then everyone reads for an hour. The kids bring their books or Kindles. I bring mine. We simply spend time reading.

This year, I added a Kobo to the mix.

I discovered that Kobo Plus is far better than Kindle Unlimited — not because of quantity, but quality. Kobo doesn’t just surface self-published books; it also includes good, popular titles from established authors and publishers. That single switch put me on a strong reading streak.

Writing followed reading.

Early in the year, I published a post on how to run OpenWebUI locally on a Mac. On its own, it accounted for around 13,000 visits this year. In total, the site crossed 44.9k page views. For years, I’ve aimed for 30k annually and never quite reached it. This year, it happened — accidentally.

Another long-standing idea finally took shape.

Over the last two decades, I’ve read more than 100 books. These books shaped how I think, how I work, and how I see faith and life. This year, I began publishing one idea from each book — not summaries, but reflections on how each book shaped me. I’ve written 17 so far. The long-term goal is 100.

Read those 100 Ideas That Shaped Me from Books I Read.

I also built.

I wrote a small Go + SQLite app called Read to Reflect, which surfaces a random saved essay for reflection. I then built another app, Zlynks, for book recommendations with affiliate links across different Amazon regions.

This second app was largely vibe-coded. The backend is in Go (Fiber), and the frontend in SvelteKit. The experience left me both impressed and cautious. AI can write a lot of code — good code — but judgment still matters. Validation still matters.

It strengthened a hypothesis I’ve been carrying.

Software may soon have its own “Substack moment” — a platform that handles hosting, payments, and distribution the way Substack did for writers.

I remain deeply optimistic.

# How I Used ChatGPT This Year

One quiet but significant change this year was how I used ChatGPT.

I used it more than any other AI tool — more than Claude, Perplexity, or anything else. Partly because it became freely accessible in India, but mostly because it fit naturally into how I think and work.

I used ChatGPT as a:

• scribe & editor
• thought partner
• counselor
• a tool for reflection

I will continue to use ChatGPT the way I use any other serious technology: intentionally, critically, and as a means — not an authority.

Read more of How I used ChatGPT in 2025.

# Talks

Every year I share what I know in conferences in the city. I also participate in "Faculty development programs". This year, I participated in a FDP of an Omani university thanks to a friend of mine. This is the first time, I did that.

You can read all the talks I have given.

This year, insights compounded through disciplined reading, reflective writing, hands-on building, and thoughtful use of AI as a thinking partner.

# Network

This wasn’t a loud networking year.

But it wasn’t empty either.

I stayed engaged with the Gravitas WINS community. Toward the end of the year, we started a structured 12-week goal-setting and weekly review cycle. That rhythm itself feels important.

Outside of formal calls, I met several community members in person. I also made it a habit to initiate simple conversations on LinkedIn — no pitch, no agenda. Just curiosity.

Nothing dramatic came out of it.

But something steady did.

This was a year of fewer connections, but deeper ones.

# Self

# Family & Homeschooling

Homeschooling continues to be one of the most meaningful parts of our life.

Josh uploaded his Grade 5 videos and is now learning keyboard. He also completed courses in HTML, CSS, and React, and built his homepage. He was visibly proud of it.

Jerry cleared his Trinity First Grade certification. He picked up guitar naturally and learned multiple songs on his own from YouTube. One of them — an Ilaiyaraaja composition — he played for my sister’s birthday. That moment stayed with me.

Both kids explored coding through Coursera. They also attended youth retreats, made new friends, went to birthday parties, and learned how to initiate conversations. The old homeschooling worry — are they becoming asocial? — quietly dissolved this year.

Josh is now doing 12th through NIOS. Conversations with a career counselor opened paths like BBA and corporate law. Nothing is fixed. Everything is exploratory.

A few years ago, my wife learned swimming at around 43. This year, she started learning to drive.

When we were in Chennai, she saw the chaotic traffic. I had been encouraging her to learn driving for a while, but she kept postponing it. After that trip, she came back and said, If I’m going to learn driving, it has to be here.

She enrolled in a driving school, went every day, and within about two months, got her license.

Now that she has her own car, it has opened up a new sense of independence. She’s already imagining all the things she wants to do on her own. Watching that confidence grow has been a quiet joy.

This year was not without anxious moments.

Josh fell sick. Twice.

The first time was during the first anniversary of my mother’s death. We were traveling, and he fell seriously ill. It escalated quickly, and we had to admit him to the emergency ward. He was on oxygen for a while. We were in Coimbatore, away from home, which made it harder.

Thankfully, with good medical care and his own immunity, he recovered quickly. He stayed in the hospital for about five days and was discharged.

The second episode came a few months later. It started as a fever, and we assumed it was seasonal. When it didn’t subside, we consulted the doctor. Initial tests showed nothing. Eventually, it turned out to be typhoid. For nearly fifteen days, he was unwell.

This time, he didn’t want to be admitted. We monitored him closely at home, gave him medicines, and cared for him day by day. Slowly, he recovered.

Looking back, the year wasn’t free of friction or fear. But in each case, things worked out — steadily, quietly, and well in the end.

This year reinforced that homeschooling is not isolation, but intentional formation.

# Fitness

For about three months, we swam daily as a family. Watching my wife learn to swim confidently — even at six feet depth— at 45 was a joy I didn’t anticipate.

The rest of the year was simpler: jogging and home workouts. I typically ran 5 km in about 45 minutes, managed one 10 km run before winter, and worked out with dumbbells and push-ups at home.

Pull-ups remain unfinished business.

Maybe next year.

Fitness this year was about showing up consistently, not chasing performance.

# Spiritual Life

Three years ago, thugs disrupted our church and forced us to vacate. Two years ago, we began thinking — tentatively — about building a church.

The congregation is made up of daily laborers and slum dwellers. They don’t have much. I had never asked people for money before. That didn't stop me from raising funds.

Friends, college mates, and well-wishers contributed. The pastor traveled extensively to raise funds. At Christmas, we dedicated the church. There is still work left — wiring, painting — but standing there felt unreal.

Read more here: Vengeance Is Not Violence.

I also preached at Jesus Calls, Chennai. Having studied at Karunya, this carried deep personal meaning.

As a Christian parent, we have always talked about the Bible, faith, and church at home. We go to church regularly. Like any parent, I wanted my sons to be baptized. But I was careful to leave the final decision with them.

One Sunday, the pastor announced that there would be a baptism and gave a call. Josh went forward on his own.

A month later, after attending the preparatory sessions, he was baptized.

That mattered to me more than the act itself — not because it happened, but because it wasn’t forced. In that sense, this year felt pivotal for his spiritual growth. I’m deeply grateful for that.

This year, faith moved from inward conviction to outward obedience.

# Travel

We began the year quietly with a weekend retreat at Shivjot Resort — no crowds, just rest.

In March, we visited Chikmagalur and stayed in Java Rain Resorts. This was a month after my elder son was admitted in emergency. So seeing him hike to Mullayyanagiri peak was heartening. The unforgettable experience was visiting "Hoysaleswara Temple". I never tire of Indian architecture.

In August, we visited Karaikudi and Rameshwaram. Karaikudi, in particular, stood out. The Chettiar approach to business, wealth, and philanthropy was deeply instructive. The architecture, food, and culture felt like stepping into another world.

We ended that trip with a few days in Chennai at MGM Beach Resort — meeting friends, talking, resting.

Travel became a way to reflect, not escape.

# Closing

In 2025, I said to my friends often, "It feels like I'm living in a dream. I am afraid I will wake up."

This is the phrase I repeated every morning before I got out of bed.

“The Lord has done great things for us, and we are filled with joy.”

That still feels true on the last day of the year.

# Retros of Past Years

I’ve been publishing annual reviews for over a decade. Each review captures the journey and milestones that shaped me. You can read them to see how far I’ve come.

Got comments? Send them to me via Twitter or join the conversation.

'Year In Review - 2025' appeared on Joseph Jude.

I used it more than any other AI tool — more than Claude, Perplexity, or anything else. Partly because it became freely accessible, but mostly because it fit naturally into how I think and work.

Primarily, ChatGPT became my scribe and editor.

I dictate a lot — while walking, while reading, sometimes while a thought is still half-formed. I ask it to rewrite in my voice, not to smooth things into generic prose. Over time, by correcting it when it drifted, it learned how I sound. That alone reduced the friction of writing. Thoughts that would earlier remain in my head now made it to the page.

Second, it became a thought partner.

Whenever I wanted to reason through something — financial decisions, product ideas, writing structure — I used ChatGPT in a ping-pong fashion. Not to be told what to think, but to sharpen my own thinking. It behaves like an encyclopedia that can narrow itself to exactly what I need at that moment.

Third, it became a tool for reflection.

At the end of each month, I would do a short retro with ChatGPT. I would dump what had happened and ask questions like: What did I do well? What should I stop doing? What patterns do you see? The answers weren’t always accurate, but they were useful starting points. Because I had been consistently adding context, the reflections were often surprisingly close.

At times, it even functioned like a counselor.

Because it knew what I had shared — my work, my faith, my habits — I could ask questions about my own tendencies, strengths, and blind spots. I don’t outsource self-knowledge to it. I still test these insights against close friends and lived reality. But it became another mirror — imperfect, but useful.

I will continue to use ChatGPT the way I use any other serious technology: intentionally, critically, and as a means — not an authority.

# Also Read

• Using ChatGPT as a Bible Study Partner
• Understanding India's 2025 Budget with ChatGPT
• Short notes on AI Economy

Got comments? Send them to me via Twitter or join the conversation.

'How I used ChatGPT in 2025' appeared on Joseph Jude.