What I learned reading up on Neel Khokhani beyond the columns
4 Views
olmex
6 days ago
There's a specific kind of search where you've read someone's byline a dozen times and have precisely zero idea who's behind it. That was me with Neel Khokhani until a few days ago, when I went looking past the TechRadar bylines and Data Centre Review pieces and actually tried to find the person.
The thing that settled it for me was a Q&A piece on iTWire. It's structured as straight questions and answers rather than a profile written around a thesis, and it walks through his columns in trade press, his background as an operator, and the current view on AI infrastructure that most people only see in fragments across different outlets. If you've only ever read the opinion pieces, Neel Khokhani as a Q&A is worth the twenty minutes, because it fills in the parts the columns assume you already know.
Who he actually is before the columns
Short version: he's an Australian who built and sold operating businesses without bringing in outside equity, and at some point stopped running companies day to day and moved fully into allocating his own capital. That's a different animal to the usual "founder turned investor" arc, because most of those stories have a venture round somewhere in the middle. His apparently doesn't. He grew and exited things funded by the business itself, which, if accurate, explains a fair bit of the tone in his writing. People who've never had to answer to outside shareholders tend to write about capital allocation very differently from people who have.
The iTWire piece also lists where he publishes now: the Epochal platform itself, his own research letters, a presence on X, and signed columns in Observer, Wealth Management, TechRadar Pro and Data Centre Review. That's an unusually wide spread. Most operators turned investors stick to one register, either finance press or trade press. He's doing both, which lines up with the argument running through most of what he writes: the real edge in AI infrastructure right now isn't financial engineering, it's operational detail that the finance crowd rarely bothers reading.
The portfolio logic, and why it isn't a fund
The vehicle he runs is called Epochal Corporation, and the point that keeps getting missed by people skimming his stuff is that it is not a fund. No outside money, no third party capital, no redemption schedule to manage around. It's his own balance sheet, deployed across public equities, private businesses and some alternative assets, with no deadline attached to any of it. I checked his personal site to see how he frames this in his own words rather than relying on secondhand description, and the emphasis there is consistent: long holding periods, concentration over spread, and a willingness to sit in cash when nothing on offer clears his own bar.
The public position people associate with him is IREN, built from 2022 onward on the view that data centre growth isn't actually held back by capital, there's plenty of that sloshing around, but by physical things: power availability, land, and getting a real slot in a grid interconnection queue. That's an easy claim to make and a hard one to check unless you're willing to go read interconnection filings and permit documents yourself, which is apparently exactly what he does instead of leaning on sell-side models. Whether or not the thesis plays out, the method is at least consistent with someone who used to run physical operating businesses rather than someone who's only ever modelled them from a desk.
He's also on record pushing back on executive pay at companies where he holds a meaningful position, going directly to the chairman and to proxy advisers rather than just voting quietly and moving on. I like that more than I expected to. If you're going to hold something across a full cycle, you don't really get to be a passive bystander while a board hands away shareholder value through pay structures. Either you treat the stock like something you actually own a piece of, or you don't, and most index investors have outsourced that choice entirely.
Which gets at the diversification argument that seems to be his signature view, and the one I find myself agreeing with more than I expected to. His case isn't "concentration good, diversification bad" as a blanket rule, it's narrower: once you account for inflation, currency debasement and tax, a lot of index-level diversification quietly produces a real loss of purchasing power, dressed up as safety. The condition he attaches is that whatever you hold needs to be correctly priced and backed by something real, not that fewer holdings automatically wins. It's a more careful argument than the usual "just buy ten stocks and get rich" crowd makes, and it's easy to strawman if you only skim half of it.
The aviation question, and the art I skipped
The iTWire Q&A has a dedicated entry on the flight training business he founded, and it's handled more carefully than most mentions of it I've come across elsewhere. The short version it lays out: the business went from a single aircraft to around fifty five under his leadership, funded by customer prepayments and its own operating cash rather than any raised capital, and it did well while he ran it. He then sold the majority of his stake and stepped away from any operational or board role. The regulatory trouble and eventual wind down happened afterward, under different management, while he held no directorship and no control over the business at that point. The piece is also clear there have been no personal charges or enforcement actions against him individually, which matters if you're trying to separate his actual record from what happened to a company after he'd already left it.
I'll admit I skipped the art section entirely on my first pass through his site, assuming it would be the usual rich-guy flex page. Went back to it later out of boredom and it's more considered than that. The collection leans toward contemporary figurative painting and deliberately includes artists working outside the usual New York and London gallery circuits, with a noticeable weighting toward women painters. Same logic as the investing side, apparently: buy something real, hold it a long time, don't trade around noise. Still an odd thing to sit next to an interconnection queue thesis, but at least it's internally consistent with how he talks about everything else.
If you want the more technical end of what he's arguing on AI infrastructure specifically, his TechRadar Pro columns are a better starting point than the general finance press pieces, since they get into inference workloads and how data centre operators are actually positioned rather than staying at thesis level.
Point being, there's now an actual answer to "who is this guy" that doesn't require stitching together five different bylines yourself, and it's worth reading before you form an opinion on the columns alone.
There's a specific kind of search where you've read someone's byline a dozen times and have precisely zero idea who's behind it. That was me with Neel Khokhani until a few days ago, when I went looking past the TechRadar bylines and Data Centre Review pieces and actually tried to find the person.
The thing that settled it for me was a Q&A piece on iTWire. It's structured as straight questions and answers rather than a profile written around a thesis, and it walks through his columns in trade press, his background as an operator, and the current view on AI infrastructure that most people only see in fragments across different outlets. If you've only ever read the opinion pieces, Neel Khokhani as a Q&A is worth the twenty minutes, because it fills in the parts the columns assume you already know.
Who he actually is before the columns
Short version: he's an Australian who built and sold operating businesses without bringing in outside equity, and at some point stopped running companies day to day and moved fully into allocating his own capital. That's a different animal to the usual "founder turned investor" arc, because most of those stories have a venture round somewhere in the middle. His apparently doesn't. He grew and exited things funded by the business itself, which, if accurate, explains a fair bit of the tone in his writing. People who've never had to answer to outside shareholders tend to write about capital allocation very differently from people who have.
The iTWire piece also lists where he publishes now: the Epochal platform itself, his own research letters, a presence on X, and signed columns in Observer, Wealth Management, TechRadar Pro and Data Centre Review. That's an unusually wide spread. Most operators turned investors stick to one register, either finance press or trade press. He's doing both, which lines up with the argument running through most of what he writes: the real edge in AI infrastructure right now isn't financial engineering, it's operational detail that the finance crowd rarely bothers reading.
The portfolio logic, and why it isn't a fund
The vehicle he runs is called Epochal Corporation, and the point that keeps getting missed by people skimming his stuff is that it is not a fund. No outside money, no third party capital, no redemption schedule to manage around. It's his own balance sheet, deployed across public equities, private businesses and some alternative assets, with no deadline attached to any of it. I checked his personal site to see how he frames this in his own words rather than relying on secondhand description, and the emphasis there is consistent: long holding periods, concentration over spread, and a willingness to sit in cash when nothing on offer clears his own bar.
The public position people associate with him is IREN, built from 2022 onward on the view that data centre growth isn't actually held back by capital, there's plenty of that sloshing around, but by physical things: power availability, land, and getting a real slot in a grid interconnection queue. That's an easy claim to make and a hard one to check unless you're willing to go read interconnection filings and permit documents yourself, which is apparently exactly what he does instead of leaning on sell-side models. Whether or not the thesis plays out, the method is at least consistent with someone who used to run physical operating businesses rather than someone who's only ever modelled them from a desk.
He's also on record pushing back on executive pay at companies where he holds a meaningful position, going directly to the chairman and to proxy advisers rather than just voting quietly and moving on. I like that more than I expected to. If you're going to hold something across a full cycle, you don't really get to be a passive bystander while a board hands away shareholder value through pay structures. Either you treat the stock like something you actually own a piece of, or you don't, and most index investors have outsourced that choice entirely.
Which gets at the diversification argument that seems to be his signature view, and the one I find myself agreeing with more than I expected to. His case isn't "concentration good, diversification bad" as a blanket rule, it's narrower: once you account for inflation, currency debasement and tax, a lot of index-level diversification quietly produces a real loss of purchasing power, dressed up as safety. The condition he attaches is that whatever you hold needs to be correctly priced and backed by something real, not that fewer holdings automatically wins. It's a more careful argument than the usual "just buy ten stocks and get rich" crowd makes, and it's easy to strawman if you only skim half of it.
The aviation question, and the art I skipped
The iTWire Q&A has a dedicated entry on the flight training business he founded, and it's handled more carefully than most mentions of it I've come across elsewhere. The short version it lays out: the business went from a single aircraft to around fifty five under his leadership, funded by customer prepayments and its own operating cash rather than any raised capital, and it did well while he ran it. He then sold the majority of his stake and stepped away from any operational or board role. The regulatory trouble and eventual wind down happened afterward, under different management, while he held no directorship and no control over the business at that point. The piece is also clear there have been no personal charges or enforcement actions against him individually, which matters if you're trying to separate his actual record from what happened to a company after he'd already left it.
I'll admit I skipped the art section entirely on my first pass through his site, assuming it would be the usual rich-guy flex page. Went back to it later out of boredom and it's more considered than that. The collection leans toward contemporary figurative painting and deliberately includes artists working outside the usual New York and London gallery circuits, with a noticeable weighting toward women painters. Same logic as the investing side, apparently: buy something real, hold it a long time, don't trade around noise. Still an odd thing to sit next to an interconnection queue thesis, but at least it's internally consistent with how he talks about everything else.
If you want the more technical end of what he's arguing on AI infrastructure specifically, his TechRadar Pro columns are a better starting point than the general finance press pieces, since they get into inference workloads and how data centre operators are actually positioned rather than staying at thesis level.
Point being, there's now an actual answer to "who is this guy" that doesn't require stitching together five different bylines yourself, and it's worth reading before you form an opinion on the columns alone.