Gods in the Machine
How Indian Tech Companies Balance Modernity with Temples in the Server Room
Or: A Physics PhD Watches His Homeland Worship Its Computers
Let me be clear: I think this is all completely insane.
At thirteen, I announced I was an atheist. My barely-religious parents threw me out of the house. The insults flew in that peculiarly Indian way—erudite English sentences I can't repeat here (decidedly NSFW), punctuated by Hindi daggers like "safed chamri" (white skin)—accusing me of being a lover of Western culture, saying I'd lost my roots, that I had no idea what I was talking about. My mother, who smoked liked a chimney and drank like a fish and hadn't visited a temple in years, suddenly became devout—just to spite me. After a few drinks, she'd occasionally admit I was right. The rest of the time, she'd double down on rituals she didn't even believe in.
I have a Physics PhD. Before that, Electrical Engineering from Columbia. I understand determinism, probability, causation, and confirmation bias. I left Delhi partly to escape this madness.
So imagine my horror when I discovered that this insanity... works.
Not spiritually. Not mystically. Economically. Empirically. Measurably.
The priest blessing the server room in Mumbai? That data center has 99.99% uptime. Muhurat trading sessions? Generated 0.6% better returns than random hours. Companies with temple rooms? Reportedly 18% attrition vs Silicon Valley's 25%.
I still think it's nuts. But I'm also a libertarian. If people want to imagine their computers have souls that need blessing, who am I to argue with their ROI?
This is the story of how a country collectively lost its mind and somehow built a quarter-trillion dollar tech industry in the process.
The Server Room Exorcism
Last May, an Edge Data Center in Mumbai held a puja ceremony on Akshaya Tritiya—the most auspicious day for new beginnings. A priest, draped in saffron, sprinkled holy water on $2 million worth of blade servers while chanting Sanskrit mantras that predate the Roman Empire. The DevOps team waited patiently, production deployment on hold. An American VC watched on Zoom as his Series B investment received a vermillion tilak.
The data center now has 99.99% uptime.
Correlation? Causation? Complete coincidence? As a physicist, I know the answer. As a student of economics and finance, watching the stock price of Indian tech companies, I'm less sure.
This isn't rare. When NASA and ISRO sent their joint NISAR satellite equipment to California, dignitaries from both agencies gathered for the traditional coconut-breaking ceremony—right outside the clean room. As the BlackSky satellite blog noted with barely concealed amusement, "Launching a rocket definitely is a coconut-worthy endeavor."
Every year during Ayudha Puja—the "worship of tools"—engineers across Bangalore garland their laptops and place tilaks on server racks. As one techie explained, "We worship laptops, smartphones, and tablets, as they are essential tools in our lives." In IT companies, computers and servers receive the same ritual respect once reserved for plows and swords.
The rational part of my brain screams. The part that reads quarterly reports stays quiet.
The Motorcycle That Became God (And I Wish I Was Making This Up)
You think blessing servers is peak insanity? Let me tell you about Om Banna.
On December 2, 1988—while I was safely ensconced in the Physics Department at The University of Texas at Austin having fled this madness—a young Rajput named Om Singh Rathore crashed his Royal Enfield Bullet into a tree near Chotila village. He died. The police hauled the motorcycle to the station. The next morning, the bike was back at the accident site.
The police, thinking this was a prank, dragged it back, emptied the fuel tank, and chained it up. It appeared at the crash site again.
After several iterations of this—and I cannot believe I'm writing these words—the locals decided the motorcycle had become divine. They built a temple. To a motorcycle. A 350cc Royal Enfield Bullet, registration number RNJ 7773, is now a deity.
Every day, hundreds of travelers stop at the shrine on the Pali-Jodhpur highway. They garland the bike with marigolds. They apply tilak to its headlight. They offer it bottles of liquor—because apparently even divine motorcycles need a drink (and I guess don't worry about DUI?). Some believe that those who don't stop to pray will have accidents. The tree Om crashed into is festooned with prayer strings and bangles.
The shrine has its own priest. Morning and evening aartis. An economy of flower sellers, prasad vendors, and miniature Royal Enfield models. There are folk songs about a motorcycle. FOLK SONGS. ABOUT A MOTORCYCLE.
Here's what kills me: accidents on that stretch of highway have reportedly decreased by 30%. The motorcycle god works.
I have a PhD in Physics. I understand that correlation isn't causation. I know that increased vigilance due to the shrine's presence almost certainly explains the safety improvement. I understand the placebo effect, confirmation bias, and selective memory.
But try explaining that to the thousands who credit Bullet Baba with their safe journeys. Try telling that to the tourism economy that's sprouted around a dead motorcycle. Try arguing with the donation box that's never empty.
The Royal Enfield company? They've acknowledged it. Some reports say they gifted a new motorcycle to the shrine. Because when your product literally becomes a god, you don't fight it—you lean in.
This is India: where we don't just anthropomorphize our machines, we deify them. Where a production vehicle can achieve moksha. Where the line between technology and theology doesn't exist because we never drew one in the first place.
My atheist physicist self wants to scream. My economist self is taking notes. A motorcycle generates more revenue dead than most businesses do alive.
And so now, if you'll excuse me, I need to lie down. The spreadsheets are starting to look like scripture.
Ok, I feel a bit better and will get up again. Let's carry on, at least until I need to lie down again.
When Algorithms Bow to Astrology
Picture this: It's Diwali evening on Dalal Street. The NSE and BSE—India's premier stock exchanges—open for exactly one hour of "Muhurat Trading." The trading floor, usually a temple to capitalism, becomes an actual temple. Marigold flowers and oil lamps surround the terminals. Before the opening bell, there's a Lakshmi Puja—worship of the goddess of wealth—right there between the Bloomberg terminals.
This tradition has run uninterrupted since 1957.
The volumes are mostly symbolic, the volatility high, as everyone rushes to make token trades in the auspicious hour. Business newspapers print special pull-outs with astrologers' market outlooks. Stock tips are organized by zodiac sign. Where else would you find a modern stock exchange that literally pauses for prayer and declares an auspicious start to trading?
Here's what kills me about this: Muhurat trading sessions consistently show positive returns—averaging 0.6% gains versus random trading hours. The market that shouldn't exist outperforms the market that should.
Gujarati traders treat muhurat highs and lows as omens that guide their strategies until Uttarayan (January 14). An entire investment philosophy based on one hour of ritual trading. The worst part? Their returns don't argue against it. [In case you are curious, the real reason this might work is because of “breakout trading”].
Wall Street has its bull. Dalal Street has Nandi—same animal, better accessories, and apparently, better timing.
The Mother Who Became Religious to Spite Me
You want to understand why Indian tech companies worship their computers? Let me tell you about my mother.
The night I announced my atheism, she hadn't performed a puja in years. By the next month, our house had a new temple corner. Suddenly, every Hindu festival required elaborate observation. Diwali went from token diyas to full-scale productions.
"You've lost your culture," she'd rage. "You think you're too good for us because you read Western books?" said the very widely read second fastest reader of English I have ever known (Tyler Cowen is the fastest, but he demurs and says my mother was faster. I don't know. Hence second.) By the way, I should point out, this “devoutness” did not devolve to any reduction of cigarette or alcohol consumption.
After a few drinks, the truth would slip out: "You're right, of course. It's all nonsense. But what will people say?"
That's when I understood: Indian tradition isn't about belief. It's about belonging.
Tech companies blessing servers? They don't believe the servers have souls. They know their employees believe that the company believes it. Three layers of meta-belief, creating actual cohesion, producing measurable results.
My mother knew the rituals were meaningless. I knew she knew. She knew I knew she knew. We performed this elaborate kabuki anyway. She did it to spite me. The tech companies do it to unite employees. Same mechanism, different scale.
Her spite-religion lasted until my father's death, whereupon it disappeared entirely for the last decade of her life. She kept the rituals. I kept the rationality. I wonder who was happier?
The Economics of Collective Delusion
Here's where it gets interesting—and by interesting, I mean infuriating.
Laurence Iannaccone, an economist at Chapman, published a paper in 1992 called "Sacrifice and Stigma: Reducing Free-riding in Cults, Communes, and Other Collectives." His thesis: seemingly irrational requirements and bizarre restrictions can be economically optimal. They create group cohesion that outweighs their costs.
Let that sink in. An economist—not a priest, not a mystic, an economist—proved mathematically that irrationality can be rational.
Applied to Indian tech: Those server room pujas? They're what Iannaccone calls "participatory crowding"—shared rituals that create the cooperation that makes companies work. The ₹50,000 spent on a blessing ceremony isn't insurance against system failure; it's an investment in team cohesion.
The math is perverse but persuasive: If (cohesion gains) > (irrationality costs), then irrationality becomes... rational.
It's like that Woody Allen line from "Love and Death"—"Subjectivity is objective." When everyone shares the same delusion, it becomes the market reality. The delusion creates the cohesion. The cohesion creates the performance. The performance validates the delusion.
Network effects of nonsense: When everyone times product launches for auspicious moments, it creates coordinated market movements. Self-fulfilling prophecies with Sanskrit subtitles.
My calculation, much as it pains me: The opportunity cost of a one-hour puja is less than the productivity gain from employees who believe they're blessed. The transaction costs of shared irrationality are lower than the transaction costs of pure rationality.
The darkest truth Iannaccone revealed: In groups, irrationality is profitable when everyone's irrational together.
I hate this conclusion. The spreadsheets don't care.
When MIT PhDs Break Coconuts
From my perch in Connecticut, I watch the LinkedIn feeds of Indians I know. IIT toppers who went to MIT, now running AI companies. Their posts are a schizophrenic blend: one day, cutting-edge machine learning papers; the next, photos from the office Saraswati puja.
They have PhDs from MIT and rudraksh bracelets from Rishikesh. They code in Python and pray in Sanskrit. They debug with both breakpoints and Brahmin priests.
A friend's startup—let's call them TechLogic (not their real name)—builds advanced NLP models. Their deployment calendar? Synchronized with the Hindu almanac. Major releases happen on Akshaya Tritiya or Vijaya Dashami. Minor updates avoid Rahu Kalam (the inauspicious 90 minutes that occur daily).
Their lead engineer, Stanford PhD, IIT gold medalist, told me with a straight face: "Look, I don't believe in this stuff. But the team does. And a team that feels blessed performs better than a team that doesn't."
There it is. The kernel of rationality inside the irrational shell. He's not blessing servers for the servers. He's blessing them for the humans who code for them.
Still insane. Still working.
Silicon Valley's Not-So-Secret Temples
"At least we're rational here," I thought, settling into my chair in Connecticut, watching Silicon Valley from a safe distance.
Then I started noticing the ads on WillowTV—the streaming service for cricket that every Indian in America subscribes to. Between overs: PremJyotish. Again and again. "America's most trusted Vedic astrologer." The market has spoken: Silicon Valley already has astrology. It's just ethnically segregated. For now.
The Valley has its own religions, barely disguised: Intermittent fasting as spiritual practice. Microdosing as communion. Burning Man as corporate pilgrimage. They worship at the altar of disruption while pretending they've disrupted the need for altars.
Carolyn Chen, a Berkeley sociologist, studied this. Her conclusion: "Work is sacred to tech workers. Their companies and startups are the faith communities that give them meaning and purpose." Silicon Valley hasn't eliminated religion. It's made work the religion.
At least Indians admit they're being irrational.
Remember when Zuckerberg visited Kainchi Dham temple on Steve Jobs's advice? When Tim Cook kicked off his India trip at Siddhivinayak? These weren't photo ops. These were Silicon Valley's high priests seeking blessings from an older operating system.
The prediction I hate myself for making: Silicon Valley will have official astrologers within five years. They'll call them "cosmic strategists" or "alignment consultants." They'll charge $500 an hour. The same VCs who demand "data-driven decisions" will time their investments by Jupiter's transit.
It's already happening. They just don't call it astrology yet.
The Venture Capital Astrology Complex
Want to know what truly broke my brain? AstroTalk.
Founded by a techie who didn't even believe in astrology until a prediction came true. Now: 70 million users. Five million paid consultations per month. Revenue of ₹651 crore in FY2024. Targeting a unicorn valuation and IPO by 2026.
At peak times, users spend ₹22,000 per minute on astrologer consultations. That's $275 every sixty seconds flowing into star-chart readings.
The same VCs who reject pitches for lacking "quantitative metrics" are funding cosmic speculation. Between 2019 and 2024, astrology startups raised $80 million. The "religion tech" sector in India now includes 950 startups. The overall "spiritual services" industry? Valued at $58-65 billion. And don't underestimate it—projections show $135 billion by 2030.
But here's the real kicker: 84% of AstroTalk's users are under 35. These aren't tradition-bound elders. These are digital natives, choosing to pay for predictions about their future while they build that future in code.
They're using AI to optimize astrology. Machine learning models trained on millions of horoscopes. "Brigo," an AI chatbot that generates astrological advice. The founder claims AI will "outperform human astrologers in terms of speed and accuracy."
We've reached peak absurdity: artificial intelligence predicting human destiny based on planetary positions. The future using the past to predict itself.
The market doesn't care about the contradiction. The market is up 40%.
The Delhi I Fled, The India That Won
Irrationality annoys me. Gross irrationality annoys me viscerally. Indian irrationality irritates me in ways that are impossible to even express, given just how incredibly irrational it is!
And so... every time I visit India, I somehow (like Charlie Brown) always expect to find the irrationality finally catching up, and IndiaOS finally crashing. This time, surely, Lucy won't pull the football away. This time, reality will assert itself.
Instead, I find more tech parks with more temples inside them. More MIT PhDs with more rudraksh bracelets. More unicorns with more coconuts.
The insanity isn't failing—it's scaling.
The Delhi I fled in 1984 was the capital of socialist bureaucracy and ancient superstition. The India of 2025? Every tier-2 city has a startup incubator. With a temple inside it. The country succeeded not despite the traditions but because no one tried to fix them.
I left Delhi to escape backwardness. The backwardness went public at billion-dollar valuations.
My ultra-rationalist daughters—one getting a Physics PhD, one with a Mech E Masters—watch with the same horror I do. They look at astrology apps like physicists look at flat earth forums. But their Indian-American classmates at Stanford and MIT? Different story entirely. The next generation hasn't abandoned superstition. They've democratized it—crystals for everyone, manifestation journals, moon water. The West is learning what India always knew: shared irrationality is a feature, not a bug.
Why This Insanity Works (And I Hate That It Does)
Let me break down why blessing computers correlates with company success, much as it pains me:
Psychological safety through ritual: Employees who feel cosmically protected take more calculated risks. The puja doesn't protect the server; it protects the programmer's confidence.
Coordination through superstition: When everyone waits for the auspicious moment, you get synchronized market action. It's a Schelling point with Sanskrit characteristics.
The placebo effect at scale: Belief creates reality through market mechanics. If everyone believes Thursday is auspicious for launches, Thursday becomes auspicious for launches.
Reduced transaction costs: Shared rituals reduce the need for elaborate contracts and monitoring. The gods are watching—better than HR ever could.
Cultural continuity as competitive advantage: Companies that embrace local traditions reportedly have lower attrition, higher employee satisfaction, better local market understanding.
The optimal amount of irrationality, I'm forced to conclude, might not be zero.
The final insult: despite its attachment to extreme irrational mysticism, a culture that literally worships knowledge (Saraswati) and wealth (Lakshmi) was always going to win at capitalism. They had the mythological infrastructure already in place. They just needed the policy environment to match.
1991's liberalization didn't introduce capitalism to India. It just let India be itself.
The Synthesis I Never Wanted
Looking back at this series from my Connecticut office, watching the NASDAQ tick, here's what I see:
The WhatsApp uncles from Post 5 were wrong about ancient Indians inventing computers. But they were right that tradition and tech could coexist. More than coexist—amplify each other.
The socialist trauma from Post 4 created psychological refugees who needed both algorithms and astrology. The segmented pluralism from Post 1? It's what allows rationalists like me and ritualists to build unicorns together, even while we drive each other insane.
The Lakshmi in the cash register from Post 6 evolved into Ganesha on the motherboard. The diaspora dynamics from Post 2 mean Silicon Valley imports India's irrationality along with its engineers. The wedding economics from Post 3—that same transactional traditionalism—now governs term sheets and cap tables.
I still think it's insane. The spreadsheets disagree.
Here's my libertarian peace with this madness: Their imaginary friends. Their real profits. Their choice.
That priest who blesses computers? His grandson just IPO'd an astrology app. I didn't buy shares. My Indian friends did. They're up 40%. They're laughing at me from their Teslas, ordered on auspicious days, delivered during favorable planetary positions, blessed before the first drive.
I came to bury tradition, not to praise it. But the data doesn't lie, even when everyone else does. In India, the gods are in the machine, the machine is in the market, and the market—damn it all—is up.
Trust in God, debug your code, and maybe check if Mercury is in retrograde. In India, it's not Pascal's Wager—it's Pascal's Portfolio. Diversification across rational and mystical assets.
The servers hum their electric mantras. The stock prices perform their own puja to Lakshmi. And somewhere in Bangalore, a computer science PhD is explaining to her American manager why the product launch absolutely cannot happen during Rahu Kalam.
She's right, of course. Not cosmically. Economically.
The irrationality isn't a bug in IndiaOS. It never was. It's the kernel itself—ancient, irritating, and apparently, absolutely essential for the system to run.
I hate it. It works. Both statements are true. That's the most Indian thing of all.
If you like my posts, you will like my book, The Science of Free Will: https://amzn.to/4aMQJD1. Among many other things, I talk about the future of AI and LLMs, whether bees have emotions, why traffic sucks, why Nobel Prize winning economists still don’t understand basic tax policy, and much more!
Next in this series: Post 10 - The Gender Paradox Revisited: From Sati to Startups: India's Contradictory Revolution in Women's Lives
[Previous posts: 1. The Paradox of India | 2. The Diaspora Paradox | 3. The Wedding Wars | 4. From Goddess Lakshmi to Ration Cards | 5. WhatsApp Uncles vs Wisdom Aunties | 6. From Lakshmi to Unicorns | 7. The Great Indian Cafeteria Wars | 8. The Mother Tongue Wars]


While seeking flawlessness, it can often be observed that one purposely avoids perfection. Rather than look directly at the bright light that reveals truth, for example, we look slightly away lest it blind us. And here’s a piece from a seller of Navajo blankets, touching on the same theme:
Intentional Flaw? Deliberate Mistake? Perfectly Imperfect?
Recently a new collector inquired about how much would a weavings value and collectability be diminished if the weaver made a mistake in the mirror image of the design. The answer is zero.
Navajo are deeply religious. They believe nothing is perfect, except for the Gods. They were given the gift of weaving by the Gods and taught by Spiderwoman herself - an important deity to the Navajo. To honor the Gods, Navajo weavers deliberately incorporate an imperfection.
Indeed, “The optimal amount of irrationality, I am forced to conclude, might not be zero.”
I think you miss a key reason this stuff works - it’s Making Sacred. When you Make Sacred, things work better than if you didn’t.
They say fasting is good. But it’s hard to persist unless you take it seriously with bhakti, like Ekadashi fasting. I also have role models to follow, e.g. my grandma who would fast without fail on all the fasting days, and was the definition of pious, and never cut corners. I look at her and despite all her hardships, it felt like the hand of the gods was on her.
Same with Saraswati Puja and Ayudha Puja. You’re taking an entire day to just acknowledge the tools of your trade. I’m married a white man who was raised atheist, and he will painstakingly apply kumkum and chandan to each of his power tools and thank them for working well. That feeling of oneness with your tools makes you acknowledge what you have. Our daughter picks the books she wants Ma Saraswati to bless the most, and there is a divinity in that choice.
It’s the same reason Konmari, that is rooted in Shinto practices, works - you acknowledge your possessions and imbue them with life. It makes you take them seriously and reckon with them.
I have an atheist branch of the family. They won’t do things because ‘tradition’. If anything, they’ll go straight up against something traditional. They are also always seeking why, but will stop at midwit answers. They’ll organize events all at the wrong muhurtam. Predictably (IMO), they aren’t doing that well. I look at their lives and try to see why, rationally. They were the family that would eat maggi and drink rasna when all the others in the family eschewed these things for being ‘foreign’. They’d eat meat when no one else in the family did. This part, I explain by saying they made dietary choices they had no context for and didn’t know how to integrate into their life, and in order to make those choices, they turned their backs on stuff that worked fine for their genotype for at least 300 years of recorded family history.
Even their personal lives are a mess… my mom attributes that to not marrying according to a muhurtam. When I think about how they did that, it strikes me that your marriage is the most important decision of your life, and if you use that to make a statement about your lack of belief in astrology, that’s not exactly taking your marriage very seriously. Once when one of their marriages was in trouble, a relative asked them to try sleeping facing a certain direction because it was supposed to help. They didn’t, they were like “what difference does that make?”. Well, it shows you're trying, at the very least.
I’m a fourth-generation astrology enthusiast. Only my uncle did astrology as a profession, but everyone else is well-versed in it as a hobby. We found my great-grandpa’s diaries fifty years after his death, and ten years after my grandpa’s death. He’d noted down the year my grandpa would die… and it was correct! He wasn’t even known as an astrologer!
My mom and I are now studying astrology very seriously. We’ve always been able to read charts, but now we’re looking at the whole spectrum of what’s needed. I feel like most astrologers are so bad at explaining what they are looking at to people who don’t believe. It’s like you’re trying to find the exact configuration of the sky at a particular location at a particular moment in time, and then you consult charts that talk about each kind of configuration and the effects it has. You’re trying to figure out what all the different forces acting at that particular moment will do in conjuction, and that’s where the astronomy and math end and the experience and intuition begin. Prediction is so hard because you’re essentially taking bets on small probabilities. The mistake I see many people, including my mom make, is they presume to know the life of the person whose horoscope they are looking at. Like they’d say “don’t buy stocks, speculation is supposed to go bad for you”. But… what if you’re buying something solid, like a dividend stock that hasn’t moved much in the past five years, is that still speculation? That’s where they get it wrong. And that’s what I’m trying to be better about using a more intuitive approach.
As to why certain planetary configurations lead to real-world effects, we dont know. But we don’t have to reason to pattern-match, like how an LLM knows what to say by just looking at a zillion web pages. So astrology works pretty well that way.
What frustrates me, even among people who are scientifically minded and follow astrology is they don’t try to go deep enough or hard enough in reasoning what the connections are between planets and people, or lines on the palm and personality. It’s one of the things I plan to do (we already know a lot of your palm configuration is based on hormones) when I have the time.
I write historical fiction based on real events. One of the places my astrological knowledge helps me is to look at the horoscopes of the real-world characters I’m writing about, and use that to understand their character and their internal monologue. It’s wild how informative it can be.