Jugakalpa Das on AI democratizing research, semi-systematic trading, and 2026 macro anomalies
Summary
Jugakalpa Das traces the convergence of macro and quant strategies in 1988 and argues that AI will lower research barriers but not eliminate human discretion in trading. He defines semi-systematic investing as machine-driven signal generation with human execution oversight to manage behavioral biases. Das highlights the importance of small, repetitive edges and robust risk management, citing his own 2012 blowup and Millennium's structure. He analyzes 2026 rates market anomalies, noting Treasuries selling off during geopolitical conflict due to inflation expectations, and advises early-career investors to focus on passion and micro-edges.
Episode sections
- 0:42 – 18:07
History of Global Macro and Quant Convergence
Das traces macro's origins to the Great Depression and John Maynard Keynes, contrasting it with Alfred Jones' 1949 hedge fund. He identifies 1988 as the pivotal year when quantitative, commodity-focused, and discretionary macro strategies converged, facilitated by the post-Bretton Woods FX market and the Greensput put.
- 18:07 – 32:47
Lessons on Fear, Greed, and Pattern Recognition
Das argues that human fear and greed are immutable drivers of market dislocation, making 100% predictive models impossible. He asserts that while rational patterns can be modeled, irrational behavior requires risk management and position sizing rather than prediction.
- 32:47 – 36:06
AI's Impact on Research vs. Trading
Das claims AI will radically lower the barriers to entry in hedge funds by democratizing research, shrinking the dispersion between institutional and retail insights. He believes this will disintegrate the oligopolistic structure of large multi-manager platforms.
- 36:06 – 40:50
Definition of Semi-Systematic Investing
Das defines semi-systematic investing as a process where 80-90% of trade signals are machine-driven (technical, quantitative, fundamental), but execution involves human discretion. He contrasts this with 100% systematic funds like Cubist or Renaissance Technologies, which he views as having higher tail-risk due to lack of human intervention.
- 40:50 – 42:58
Semi-systematic investing and human oversight
Speaker_2 defines FRGM Capital's approach as semi-systematic, where machine-driven signals account for roughly 90% of the process, but human traders must vet and execute every trade. This structure ensures human oversight during market anomalies that algorithms might misinterpret.
- 42:58 – 53:12
Lessons from 2012 blowout and durable edges
Speaker_2 recounts going broke in 2012 due to inadequate hedging in oil spreads, emphasizing the need for scenario analysis. He argues that successful funds like Renaissance Technologies and Millennium started with small, repetitive, micro-edges that were later scaled.
- 53:12 – 1:05:15
AI's impact on research versus trading
Speaker_2 asserts that AI significantly reduces the margin of error in research tasks like pattern recognition and data analysis, but cannot replicate discretionary chart reading or capture fear and greed. He believes AI lowers barriers to entry for research but does not eliminate the need for human strategic integration.
- 1:05:15 – 1:17:08
FR Global Macro and FRGM Capital symbiosis
Speaker_2 explains the relationship between his research firm (FR Global Macro) and hedge fund (FRGM Capital). The research unit provides market intelligence and historical comparisons to the fund, while the fund's trading needs drive the research's quantitative and fundamental development, operating under strict Chinese walls.
- 1:17:08 – 1:20:02
2026 Rates Market Dynamics and Geopolitical Anomaly
The speaker analyzes the 2026 US rates market, noting a rally in February driven by AI fears and tech sector weakness, followed by a sharp sell-off after geopolitical escalation in late February. He highlights the anomaly of Treasuries selling off during a Middle East war, contrary to historical safe-haven behavior, attributing it to inflationary war expectations and AI Capex narratives.
- 1:20:02 – 1:26:08
Macro Hedge Fund Performance and Positioning
The speaker notes significant blowups and distress in the macro hedge fund space, citing Tola and Cyril, and describes extreme positional bias and pain stretches in rates. He contrasts the poor performance of risk parity (RV) and discretionary macro funds with the relative consistency of systematic funds after initial volatility.
- 1:26:08 – 1:29:31
Market Structure and Real Money Behavior
The speaker explains that 70-80% of macro flows come from price-insensitive real money (pensions, insurance), while speculation drives short-term volatility. He argues that durable edges are rare in macro due to this structure and that the market is currently overstretched, with risk premia being heteroskedastic.
- 1:29:31 – 1:36:31
Career Advice for Early-Career Investors
The speaker advises students to identify their passion and strengths early, using Michael Phelps as an analogy for continuous, uncorrelated improvement. He suggests that AI lowers barriers to entry, allowing individuals to build independent careers if they can innovate and find micro-edges through deep passion and understanding.
Assets discussed
| Asset | Kind | Stance | Context |
|---|---|---|---|
| Crude oil | commodity | neutral | I remember 2012 and all, I used to trade oil, and me and my friend we used to be in. |
“Stance” is how the speakers talked about the asset, not a recommendation.
Claims worth verifying
FRGM Capital is 90% machine-driven but requires human execution for every trade.
Unidentified speaker (speaker_2) ·42:20
How to verify: Check FRGM Capital's marketing materials, pitch decks, or regulatory filings (e.g., Form ADV) for their stated investment process and risk management protocols. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Primary sourceSpeaker_2 went broke in 2012 trading oil spreads and had to return to the sell side.
Unidentified speaker (speaker_2) ·43:20
How to verify: Verify through Speaker_2's LinkedIn profile, professional biography, or press releases from his subsequent roles at sell-side firms. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Renaissance Technologies made 80% profit in 1988.
Unidentified speaker (speaker_2) ·45:37
How to verify: Check historical performance data for Renaissance Technologies' Medallion Fund from 2026 via primary sources like Gregory Zuckerman's book 'The Greatest Trade Ever' or fund performance archives. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Millennium Management had only one year of drawdown in 36 years.
Unidentified speaker (speaker_2) ·47:38
How to verify: Verify Millennium Management's annual performance records from 2026 to 2026 via their official website, press releases, or third-party hedge fund data providers like HFR or Preqin. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
FRGM Capital was started three years ago.
Unidentified speaker (speaker_2) ·44:19
How to verify: Check the SEC EDGAR database for FRGM Capital's Form ADV filing to confirm the date of registration or formation. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Primary sourceStan Rakamilla is a great chartist.
Unidentified speaker (speaker_2) ·53:52
How to verify: Verify Stan Rakamilla's professional background and reputation in technical analysis through industry publications, CMT Association records, or his own professional profiles. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Speaker_2 is 41 years old.
Unidentified speaker (speaker_2) ·1:13:41
How to verify: Verify Speaker_2's age through his public biography, LinkedIn profile, or press interviews. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
FR Global Macro is a pioneer in Elliott Wave analysis for US Treasuries.
Unidentified speaker (speaker_2) ·1:08:10
How to verify: Check FR Global Macro's research publications, white papers, or marketing materials for claims regarding their Elliott Wave methodology and coverage of US Treasuries. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
FR Global Macro does not give trade ideas but provides market intelligence.
Unidentified speaker (speaker_2) ·1:15:57
How to verify: Review FR Global Macro's research reports or client communications to confirm whether they provide explicit trade recommendations or only analytical commentary. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Speaker_2's previous fund, Oak Ray Capital, failed to survive in 2012.
Unidentified speaker (speaker_2) ·52:03
How to verify: Search for news articles, press releases, or regulatory notices regarding the closure or liquidation of Oak Ray Capital around 2026. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
January 2026 saw big corporate issuance and bonds selling off, while February saw a blockbuster NFP on Feb 5th that the market faded, leading to a bond rally.
Unidentified speaker (speaker_2) ·1:17:28
How to verify: Check US Treasury yield data and NFP release dates for Jan/Feb 2026 via Federal Reserve H.15 or BLS datasets. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
In mid-February 2026, 30-year Treasury yields fell to 385 basis points due to AI fears and a Citron Research article on disinflation.
Unidentified speaker (speaker_2) ·1:17:48
How to verify: Check 30-year Treasury yield history for Feb 2026 via Bloomberg or Treasury.gov; search for Citron Research articles on disinflation from that period. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
On the Sunday following late February 2026, 30-year Treasury yields peaked at 385 basis points before rising to 535 basis points.
Unidentified speaker (speaker_2) ·1:18:08
How to verify: Check 30-year Treasury yield daily data for late Feb/early March 2026 via Bloomberg or Treasury.gov. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
A Middle East war in 2026 caused Treasuries to sell off, contrary to historical safe-haven behavior.
Unidentified speaker (speaker_2) ·1:18:41
How to verify: Correlate geopolitical news headlines regarding Middle East conflicts in 2026 with US Treasury yield movements via news archives and market data. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Jeffrey Gundlach predicted the 30-year Treasury rally (or movement) correctly.
Unidentified speaker (speaker_2) ·1:19:15
How to verify: Search for Jeffrey Gundlach's public commentary on 30-year Treasuries in early 2026 via Bloomberg or his firm's (Bridgeway/Gundlach) reports. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Tola and Cyril macro hedge funds blew up or were crushed during the 2026 volatility.
Unidentified speaker (UNKNOWN) ·1:20:02
How to verify: Check hedge fund liquidation notices, SEC filings, or industry news reports for 'Tola' and 'Cyril' funds in 2026. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Primary sourceBluecrest risk parity funds were killed in February/March 2026.
Unidentified speaker (speaker_2) ·1:21:43
How to verify: Check Bluecrest Capital Management performance reports or news for Q1 2026. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Real money investors (pensions/insurance) are not buying 30-year bonds until yields reach 6%.
Unidentified speaker (speaker_2) ·1:24:17
How to verify: Survey institutional investor sentiment via surveys like the Federal Reserve's Senior Loan Officer Opinion Survey or institutional bond dealer reports. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Michael Phelps finished fifth in the Sydney 2000 Olympics and was crushed by Ian Thorpe in Athens 2004.
Unidentified speaker (speaker_2) ·1:29:51
How to verify: Check official Olympic results for swimming events in Sydney 2026 and Athens 2026 via IOC archives. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Paul Tudor Jones attended the University of Virginia and Jeff Yass attended Reed College.
Unidentified speaker (speaker_2) ·1:36:11
How to verify: Check biographical information for Paul Tudor Jones and Jeff Yass via their firm's website or reputable biographies. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.
Do your own research
- Verify the claim that 1988 was the pivotal year for macro/quant convergence by reviewing historical hedge fund performance data and market structure changes post-Bretton Woods.
- Check the performance and risk metrics of Millennium Partners' flagship funds to validate the described risk management structure (50% drawdown cut, 7% exit).
- Review oil spread correlation data from 2012 to understand the specific model failure described.
- Analyze 2026 US Treasury yield movements against geopolitical news headlines to assess the validity of the 'inflationary war' thesis versus safe-haven demand.
- Research the operational structure of FRGM Capital and FR Global Macro to verify the 'Chinese walls' and symbiotic relationship described.
- Look for recent blowups or distress in macro hedge funds (e.g., Tola, Cyril) to corroborate the claim of significant industry distress.
- Compare the error rates and output quality of AI research tools against human analysts in pattern recognition tasks using independent benchmarks.
Full transcriptexpand
0:13Unidentified speaker (UNKNOWN): In
0:13Unidentified speaker (speaker_0): conversation with Jugakalpadas, FRGM Capital.
0:17Unidentified speaker (UNKNOWN): So
0:18Unidentified speaker (speaker_1): Das, welcome to One Basis Point. It's a pleasure for us to have you on today. I guess the first thing that I want to ask is: how has Global Macro evolved from its early description? discretionary and systematic managers to today's multi-manager platforms and how has that changed the way macro risk is taken and managed? Hey
0:35Unidentified speaker (speaker_2): Krishnrahan, thanks. Thanks for inviting me. Yeah, I think that's a great question to start.
0:42Unidentified speaker (UNKNOWN): Uh
0:42Unidentified speaker (speaker_2): if you really look into what happened in the last 200 to 300 years from the Dutch to the English to the Americans who have ruled capitalism over time, finance as an overall stream is very localized on that broader scheme. And macro is not that localized, and people don't realize that part. Like finance came after macro came, if you really see big picture. Global macro started around the Great Depression. times and a little before that through so John Maynard King's the famous King's College. So they are probably the pioneers of this so-called hedge fund. In paper and theory, Alfred Jones, 1949, started the first hedge fund where he had a bunch of hundred investors and went for 20% cut and fees. But if you really look into global macro, global macro started back from that side of the water through Kings.
1:48Unidentified speaker (UNKNOWN): I
1:48Unidentified speaker (speaker_2): believe people, when they come to finance, they start with the cult book, Reminiscence of a Stock Operator, Jesse Livermore, 1929, and how everything moved to that. People should also look into how John Maynard King's navigated that time. So JP and John Maynard Kings were on the other side of where Livermore was on the was one of the greatest short sellers at that time. Now, fast forward into the 1950s, 1960s, where Alfred Jones started, but there was no concept of FX that time because it was Brettonwood's Brettonwood system. and most of it was dollar pegged. 1971, when Brettonwood system broke, then FX market became active. And if you see, that's the time where there was a smooth transition of quantitative, discretionary, both kinds of macros coming into the world. Now we can talk from both spaces. People in today's world talk about four asset classes, right? Commodities, FX, rates, and equities. But investments and over the long period of time were mostly smart stock pricking, or which you call like a long-term equity investment. After Bretton
3:07Wood, and that's the time when FX became a very important piece of it, FX market became active after that. Banks started becoming a part. of it. Now let's go into how macro transformed through this period. So Edward Thorpe was the forefather of quantitative trading. And he started Princeton Partners around the time 1970s. And uh Princeton Partner successfully ran money till 1988, where he was in the Rudy Giuliani problem with Michael Milcom and all, and they had to shut down. Till date, one of the best returns the world has ever seen, and you can compare them to Rentag, D Shaw, every other consisted quantitative units there. Well, that's where the quantitative part of trading and hedge. funds has come. Now, if you look into macro on the other side, you can divide macro into commodity focused global macro
4:12and discretionary global macro. Now, let's first talk about commodity focused global macro. Commodity focused global macro was basically from the pits of Chicago and a unique place
4:25Unidentified speaker (UNKNOWN): in
4:26Unidentified speaker (speaker_2): Princeton. New Jersey called Commodity Corporation. So MIT guy, Helmut Maymar, collaborated with a very senior cotton trader, Amos Hosteter, and the first Nobel laureate Paul Samelson and started Commodity Corporation. Now, the big understanding of multi-manager concept, all of it started in the 1970s through Commodity Corporation, if you really see. And if you see the greatest trend followers who have really made most amount of money till the 1990s all came from the DNA of CC.
5:06Unidentified speaker (UNKNOWN): So
5:06Unidentified speaker (speaker_2): CC has given birth to people like Bruce Bruce Kovner, Covner, Michael Michael Milkham, Milkman,
5:12Unidentified speaker (UNKNOWN): Ed
5:12Unidentified speaker (speaker_2): Ed Sicota, Sicota, Louise Louis Bacon. Bacon. So these were like Caxton, More Capital, Pound Terra. So pretty much the commodity generation has come through the commodity pit in which you had people like uh Richard Dennis, turtle traders, and uh you have the other CC people who really made discretionary. trend following to be the money-making machine. And coincidentally, after Bretton Wood, throughout this time, commodities had one of the biggest trending markets. And both of them fall in the same page. Now, by the way, Paul Truder Jones. Paul Truder Jones started with Martin Malcolm and then he went into that 1987. He just crushed it there. So, but on the other hand. side, which is not commodities global macro, which is discretionary global macro, you can pretty much divide it into starting with
6:12uh George Soros and Jim Rogers, and from there Julia Robertson, and Soros Rogers, Julia Robertson, and 1987 Paul Dr. Jones. They basically covered because ERM came around 1980s to 1919s. FX became one of the greatest places to put into. And people like Stan Brackemiller and Soros, they made substantial money in 1992 ERM into the Stalin crash and all. So these are discretionary. So it was way more oligopolic there. Now, to me personally, the most fulcrum year of global macro is 1988. And 1988 is the time where all these three places of quantitative macro commodities, global macro, and discretionary macro, three of them came together. Uh Alan Greenspan became the Fed chair in 1987.
7:16His welcome note was October 19th, Black Monday, and 30th October, he made the first rate cut called the famous Greenspan put, how it started. Now, that's the time when the central banks had a paradigm shift. What I call as the move from Austrian Australian economics to keep Keynesian economics, which is like central banks trying to be the cushion to float and grow the overall base of money. So why 1988 is so important? By the way, uh Sean Phil, one of the pioneer multi-managers, they came in 1989. Sean Phil Fin 1988, Millennium 1989. But they were like very young there. And Millennium came after Easy Englander went through two other attempts of startups and it didn't work. So the multi-manager platform started around that time. Now let's uh let's go to what happened on 1987 on that day. 1987, on the Black Monday, Soros lost 300 million. Paul Tudor Jones became
8:27the talk of the town. Paul Tudor Jones and Peter Boris somehow showed the world that technical analysis, if you go into a kind of a projected pattern recognition to 1929, they predicted the stock market crash. And uh discretionary macro was getting hammered there. But at the same time, a bunch of other people started making substantial money there. So at this time, through this period, the most important bank that really supported this entire infrastructure of 1985 to 1990, where all this paradigm shift was coming, is basically Morgan Stanley. And Morgan Stanley was basically called the quant house of the world. So we can divide this beautiful story into three histories coming together and they confluence in 1988, 89, and then you get
9:361992, where in the FX markets Soros and Quantum made a billion dollars in one day on Deutsche Mark versus Cable, and then comes 1998 where LTC MD Battle came, and then comes dot com, and then you come to modern era. Now I would try to emphasize five minutes on this quantitative. uh history into 1988 and how discretionary macro overlapped there. Now think about it. Uh so after Edward Thorpe, which I explained to you guys, where he was the forefather of quantitative investing and he was doing great in Princeton Partners, there was not a lot of quant funds who were interested and personal computing was slowly coming into the picture in the late 1980s, right? So Simons was one of the pioneers in this space. So Simons initially started collaborating with Lenny Bam, the famous Bam Wish
10:44algorithm guy, uh codebreaker from IDE. Like they used to work to together to break Russian codes in Institute of Defense Analysis, IDA. Then with Lenny Bam, he was doing discretionary trading, a lot of FX, because after 1971 to 1980, that's the time FX market was really good. They used to have this tape, and even through tape you can information train trend the market there. And that's the time when especially guys like Bruce Kovner and all commodity corporation trend followers were also making money, the 1970s part. Now then Simons and Lenny Baum didn't go along well. And then he moved out of that. And uh then he started with a guy named James Axe, who was his previous colleague from uh Storybroke. Stony Broke. Uh Axe is famous for axe quotient and all. very theoretical guy. Developed one of the first pioneer factor-based kind of trend following kind of a model for them. But that didn't work. And in 1987 and all, Simonson Axe broke out and he collaborated with a guy named Yvelyn Evelyn Burlicam, who was a
11:59commuting communitarian, and uh he started. doing short-term stuff there. Now that's the same time where the relative value concept of today's trading also started by a guy named uh Bamberger. So he was one of the quants in Morgan Stanley. And in 1985 he started, 1986-87. he got a boss named Nanzia Tartalia, and they were doing something. So they started a unit called uh PET, uh, which is basically Morgan Stanley's quantitative unit where they're doing pairs trading. So Bamberger Berger was the first guy who identified that you go into the world of equity. equities and figure out any two correlated stuffs and try to trade the mean reversions of those statistically correlated units. It can be PepsiCola, it can be anything else, right? So, and but they were a senseide bank, Morgan Stanley. And but then PDT, imagine, gave birth to two of the greatest founders of quantitative trading. They hired a guy named David Shaw. And then so David
13:19Shaw in 1986-87 is the time, I guess, is where Shaw Shaw Shaw basically went to them. And then there was a guy named Frey. So Simons wanted to crack into this kind of trading, which in today's world. in the big picture is RV. If you really see big picture, how relative value trading has come. And uh Robert Frey, David Shaw, and Bamberger, these three were the pioneers of what you call as correlation trading and payer peer trading. And Morgan Tanley under APT. is like the house where this concept started. But that APT unit broke in 1988. And that's the time when Robert Frey moved with Simons into Rentech. And David Shaw got funded by another hedge fund guy called Donald Sassman, Palona Partners. He started D. Shaw. Bamberger started collaborating with Edward Thorpe at Princeton Partner. So I want us to start the discussion of really
14:37understanding how every world got together in 1988 and the world of quant investment and macro all started around that. Because in today's world, even in macro, our RV is one of the biggest parts of it. And the concept of overall RV, where you hedge in first principle component or second principal component, or third principal component, everything of that, the first guys who really thought about it came from that. And 1997, 97 Commodity Corporation got bought by Goldman Sachs and through the 1990s and all. the trend following came. Then comes 2000 dot com bubble. And by that time, Millennium really killed it because Millennium were the first guys who made this multi-manager model, and Easy Englander just funded pods and pods and pods. And 90 and then a lot of times when everyone else was drying out, Millennium was so hedged and
15:37they consistently made money. Sean Field was like the old school places where people were coming and like BAM came out of Schoenfield. Dimitri Balia's near all came out of Schoenfield on that time. And then quantitative trading went into a little more dimension because David Shaw grew way more than Simons in the 1988 to 90. But then Simons went when because Evelyn Burlicam left Simons in 1988, then Robert Mercer and Brown they joined from IBM with Simons, and then they deployed very similar strategies, which are so Burlicam was one of the first guys who involved technical trading into quantitative unit that time. And uh that's the time when Rentec invested into equities and then. it was a Cinderella run for them. Rentec just killed it through that time. And not to forget to mention Saskena came very similar times. And from Saskena, Jane Street came later on. So that body moved into pure quantitative unit. Uh, Millennium and Schoenfield Sean Field gave birth to uh the Susan. school of multi manager parts because 2000 what happened? Two of the
16:56biggest discretionary parts, which is uh Tiger, Julia Robertson, and Quantum, both of them broke out because dot com they just got sloshed. So multi-manager parts got all the talents coming Came out. Tigers disintegrated, and then came the lights of Stevie Cohen. uh Chris Shiroko's, Brevin Howard's, and all through that time, and then everything moved into multi-manager on this versus a school of quantitative guys. But both of them started doing very similar on the concept but different on execution kind of style work, and that's where we come into today. So that's like a pretty clean understanding of how finance evolution of hedge fund has come throughout?
17:44Unidentified speaker (speaker_3): So that was actually a really beautiful perspective and comprehensive history on on how macro trading has evolved. Following up to that, what lessons from early generations of macromanagers Managers still apply and which of them no longer do? What industry developments most influenced how you built FR Global Micro? And what do you deliberately avoid?
18:07Unidentified speaker (speaker_2): It's a very broad-based question, and it's something we should all think about as being money managers and founders. Like I always believe in two things. One is history always rhymes. And you cannot think of another color other than the seven colors of VibTure. That's a human capacity. You cannot go and out. So supervised learning is the only thing that's there at your disposal. You can only forecast something which has happened or which can be predicted. And I'm saying that standing today in this agentic world. So that's why for me history is very important. How things have come. Now, to your first question, that what works and what doesn't work. I think first of all, as humans, we should really defocus and go 10,000 years back. What are we talking about? We are talking about 50 to 60 years of let's say pattern recognition in an evolution of a race, which has been for so many years, right? DALLEO Dalio started an attempt in 1982 and all, didn't work out. He actually worked. with Paul Tudor-Jones to start something called all-weather strategy. Paul Tudor
19:20gave him access to the research during 1986 and all. He wasn't able to give good sharp. Dalio told, Can I take it and go away? Paul is like fine. 1987, Paul made super money. Dalio also made some money, and he started getting funded by some of the pension funds in UK, and then it became history because DALU, that's the time when he started rule-based all weather model. And then he somehow grew into the biggest hedge fund in fundamentalslash rule-based conceptual strategies. But his growth was mostly from the 1990s, and that all weather really spiked because he really killed through the dot-com 2008, he really strong to that. And from an entrepreneur entrepreneurial perspective, if you really focus on the work with Edward has done and which went, Citadel has to be mentioned, 1991 and all, where Ken started from such a garage back in Harvard, then he really was amazing in hiring the smartest people. Like Ken's mentality was very similar to
20:23what someone at Palo Alto or Silicon Valley is like here. He would just go to the smartest bank, give the most amount of money, and hire the smartest people and make them option pricing into slices and dices and make it better. So given though this thing let me jump back to your point of what works what is not what doesn't work. Let's defocus back and understand that human fear and greed will never change. So there are there are two aspects of other races versus humans. Humans are superior than every other mammal species because we think rationally. That is our difference from let's say a monkey. A monkey sometimes doesn't think rationally, but humans, on an average, most of the time think rationally. But there are certain times where fear and greed into humans takes a toll and humans start thinking. irrationally. Now, if you really think from that point, apply it to finance, apply it to anywhere else. Can you predict and model the irrational part of it? The answer is no. But can you predict and model the rational aspect of it? The answer is yes. So, what works and what doesn't work? I would say, like, I'm a great follower of Howard Mark and from that perspective, I would say that it's a gray area. I don't think
21:43anyone of us, even in this 50 years of finance, can understand what works versus not, versus 500 years of human evolution, we cannot w know what works versus not. The only thing that we know is there's There is a substantial amount of time that rational thinking is there. And there are times where. human beings' brain goes into fear and greed where the rationality goes on a toss. Now, what we actually do, we look into different factors all across us, and we want to pattern recognize what's happening to the output of anything which is based on human emotion. There is nothing that we do in finance is. more than pattern recognizing and anticipating what the other part of the unit is going to do. And you want to have something like a front run into that. That's that's that's broad-based what finance is, if you really see. So in that case, the only thing that works is we should all know that humans have fear and greed. And everything else might not work. And later half of this talk, we can go into how AI can change everything, what can happen. But this is a very important thing which happened through the journey of personal computing during
23:03the 1980s and 90s, where quantitative hedge funds started. And also today, in the world of AI and all, this concept of fear and greed, which has been 500 years back, will also be there. will be in that inflection point versus 2026 where AI becomes the inflection point. We should all know that human fear and greed cannot be overpowered by anything, and there is nothing more than supervised learning. So that's your answer: what works versus what not. I believe nothing in the world is absolutely true that works. There is nothing no hedge fund, no allocator, or no portfolio manager can ever tell you that he or she is 100% sure that this works. There is always a tail race to that. Nase Mtaleev has done and Mark Spitznegel has done amazing work on that. It can be something which has historically happened or something which is first of its time. I would say.
24:03Tiger and Quantum were the two biggest funds, and both of them got absolutely crushed in 2000 and could never come back. So in 20 years of their success, making billions of dollars through the best period of computation coming into this world, don't you think they have created scenario analysis and patterns to have saved themselves? Of course, but human fear and greed is always there. Soros lost 300 million in 1987. Soros made 1.5 billion in a day with Stan in 1992. 2000, while they were getting,
24:44Unidentified speaker (UNKNOWN): they
24:44Unidentified speaker (speaker_2): lost around a billion. You never know. They thought they might have come back. So all these things are fear and greed. in that point. So to summarize, this is a great question. Is maybe me and you will talk in 30 years. And we will come back to the same question that what works versus what not works. Now, if you deso it will be just 80 years of 500 years of evolution. The only thing that we should know is nothing works forever. And the only thing we can anticipate that at the end of the day, everything is revolving humans. And humans have time when they become irrational. Now, I will point out a very interesting story here for you guys. Finance is based on today's world, finance is more based on what fundamental and quantitative analysis is, right? But imagine entire finance has come from technical analysis. Like in the 1500s, Christopher Kurt of Nuremberg was one of the first guys who started understanding pattern recognition in
25:51rice and barley markets, bartered system. So he was also doing pattern recognition of human fear and greed there. And in the 1700s, there was the this Japanese rice merchant Monahisa Hama. He started this open high-low chart of candlestick. So that history is 300 years old. I just told you guys the first 30 minutes the history of this trillion dollars of industry, which is what 60 years old? And now I refer to Monalisa Monanli Sahama, Hama candlestick pattern. Japanese rice merchants who were doing it in 1700, 300 years old. So the pattern recognition started way back. You might be thinking today that I'm the smartest guy in this evolution at that. You have no guarantee that in 50 years someone else will think they are smarter than you, right? Because it's a marginal development in your head, right? But the patterns are a reflection of human emotions of fear and greed. There's nothing more than every pattern, because at the end of the day, what is a market? A market is a place where human emotions converge. And I can go into the world of Paul Simonson, Eugene Fama of efficient market theory, where they say that everything is priced in the market. But if
27:11you see the real money over the world is made by the school which does not believe in efficient market theory. Because the commodity corporation guys, they were all trend followers. So it was not efficient market theory, market has dislocation. And why does market have dislocation? Because human fear and greed come. And when human fear and greed comes, rationality goes for a toss. And when rationality goes for a toss, any weird thing can happen. You can think three monkeys are sitting in a room and trying to trade. Ed Sicota, one of the greatest guys, he told that some traders love to lose money. And the first time I read about Ed, I was like, what do you mean? Humans never want to lose money. Humans also want to make money. That's a decision that you did. But Edward Thorpe told that when he was playing poker in Vegas, he would say that. some poker gamblers, let's say a poker gambler has lost $10,000. The marginal benefit of making hundred
28:11dollars for him there is meaningless. He would do stupid things, he would have no emotion of winning or losing the next hundred dollars. That's called irrationality. And markets will show you those patterns of irrationality. And today you can say quantitative world, you can do higher degree kernel regression, pattern recognition, neural network, or any kind of agentic way of pattern recognizing a flow, a trend. But Rolf Elliott, who was an accountant in the 1950s, started a technical pattern recognition called Elliot Elite Wave Theory.
28:49Unidentified speaker (UNKNOWN): All
28:50Unidentified speaker (speaker_2): our portfolio. managers, it works amazing. It has three impulsive waves, two corrective waves. It is a great science, EWT, CMTs use it. And the patent recognition that you can see on that sometimes has better statistical numbers than Then if you put so much econometrics into pattern recognition. That's because firstly. it's an art, it's not a science, because there will be type 1, type 2 errors. So you really cannot see what 100% work, nothing 100% work. But pattern recognition has to think that a lot of times rationality goes for a toss and things go noisy and weird. And things can go extreme. Heteroschedasticity can come into it. It can be like the market is not moving, moving, moving. Boom, see what happened on Thursday, Friday in equities. Bloodbat. Now, can you hundred percent recognize these patterns and be a money-making machine? The answer is no. I would not be doing it in 2026. No one will be able to do it in 2056. No one will be do no one was able to do it in 1700. Christopher Kurt wasn't stupid. How can you prove Mona Lisa Hama was stupid? Christopher Kurt was
30:09stupid, Rockefellers were not there. How can you two Keynes never thought about it? You just don't know, right? So this is life. We are 99% rational, and everything that we predict to make money out of the zero sum game is under the assumption that the world is rational. But there are times where fear and greed comes, when rationality goes for a toss, everything goes into in our stochastic world you call like RW. And when that that happens, there is no way you can recognize any patterns. It's like a bunch of monkeys in a jungle. How can you pattern recognize that, right? And no quantitative fund or no technical fund or no macro funds or anything can do. All you can do is like size properly so that you have enough chip to bet on the next time when rationality comes back. So in a big picture, that's what money management is.
31:05Unidentified speaker (speaker_1): Now that makes complete sense. And I love how you put that people who don't believe in efficient market hypothesis are the ones who actually end up making money.
31:15Unidentified speaker (UNKNOWN): I
31:15Unidentified speaker (speaker_2): mean, I would not say that's 100% true because Commodity Corporation, one of the partners, was Paul Samuelson, and he was the biggest efficient market theorist. Uh, people have made money by convertible warrants at Torp, which is under efficient market theory. RV trading comes under efficient market theory versus trend following and all comes under the other school, I would say. Simon's Rentec were complete on the other side of efficient market theory. So there are there are two parts. One is called beta, and there's something called alpha. The beta part of what in hedge funds we talked about comes under that market prices things properly. Beta is like they would make as much money as the whole stock market makes. Now, the school, which is the different side where technical analysis. and discretionary macro and a lot of time people come is the school which says market is not efficient. There is risk premium in the market, and things can dislocate for a long time and doesn't converge back. And that's the place where smart people or people who can analyze. The smart is a very subjective word. I would not use it, but people who can analyze analyst can jump back and say it. So that ultimately. alpha part, if you see from nineteen seventies to two thousand crash, is mostly made by the people who are not from the
32:34efficient market theory school.
32:36Unidentified speaker (speaker_1): Okay. Yeah, yeah. That actually makes more sense. The most natural follow up I have is where do you see the industry heading ahead? Like do you think there's more consolidation coming in or more specialization, or is this gonna going be to the start of something else?
32:47Unidentified speaker (speaker_2): Well, every time we think it's the start of something else. But yeah, so there is no doubt that uh AI has radically changed the last 20 years of quantitative research, fundamental research, technical research, and monetizing money out of that research. Because I would say AI's uh impact into research is way higher than. AI's impact into trading decisions. So AI over the time will make research very, very available to a lot of bigger investment area and the discretion between institutional research versus retail that that dispersion will shrink. Because my understanding was like economy of scale and scope will make research available to broader people and the
33:54cost will reduce if if I'm look looking on the right side of where AI can go. What that would do is it would paradigm shift. shift the, I would say, the oligopolic nature of hedge fund structure, which is like, and that's I think this is a very sweet spot where I can say how FRGM and uh new hedge funds will have an edge, is a lot of things which would have needed a lot of funding and collaboration and years of experience. and experienced hires to really build up research and then to monetize that into trading. In today's world, that might be done faster. So you can think of it way similar to a couple of smart dudes starting a startup in Palo Alto Bay Area. That kind of story is coming into Wall Street because. research can be done by AI and others with very less experienced people. Now, I'm not telling you that experience doesn't have value. The prompt has to be done by the people who are experienced, and that's where the entire value is. But over time,
35:08that will also normalize. And I'm saying it's going so fast. So maybe over the next two years that'll get normalized. there. So from that point of view, I would say the biggest paradigm shift that will come is the industry will get super disintegrated. And what that means is the big uh barriers to entry that we face today against the Ballyes and the Millenniums and the 0.72s will get smaller and smaller and it will be m way more disintegrated. in our space. And that is a huge shift which probably has come through AI.
35:46Unidentified speaker (speaker_3): Yeah. AI has revolutionized research in in a big way. And in science especially. So so moving on, I've I've really got to ask um and since more moving towards the edge side of things, what is does semi-systematic semi-systematic investment? investing mean in practice at FR Global Macro? How do you decide which parts of the investment process should be systematic versus discretionary? And can you give an example where the two of these uh disagreed and you had to decide which one to trust?
36:17Unidentified speaker (speaker_2): Yeah. I think the word semi-systematic is honestly to be on both sides of the story. Like if you want to put a discretionary bear. bet, you put a discretionary bet. If you want to put a systematic bet, you put a systematic bet. Allocators come and you say, Oh, I'm semi-systematic. Jokes apart. But what I'm trying to say is, let's go back into 1988. That's the time when I always believe systematic trading really came and it killed. Performance-wide, Performance wide, it just killed like D Shop, Rentec. Saskana, Jane Street, and all they just killed. But at the same time, discretionary macro also grew through the same time. So I would say there is not a lot of firms in this world who can say we are 100% systematic. That's a very strong statement. 100% systematic is like you would never put a trade yourself, it'll be every time machine driven. If you have a fight with your wife, doesn't impact the trade. If you have a great day, doesn't impact the trade. If you have two more shots over a dinner, doesn't impact the next EPM. Now, in my opinion, Cubist right now is doing very similar stuff. Rentac over the years, that's the thing that you
37:31can backtest without a bias, where executions are all signal generated. and there is no there is zero human involvement. Now if Rahan you asked me is that the holy grave? No. Because flash crash 2008 you can blow out and when and the part is like bigger part of the market flows are discretionary. So that can easily kill you on that. Your edge will be there but there will be some time when there'll be a blowout. and you will feel like the payoff will feel like selling insurance, and then one day you blow out. That will happen. But that's what I would really define as 100% systematic. Now, my background, now which we haven't mentioned yet, I come from a day trading background where I was a The chartist, big technical guy. So from year 19 till 30. for the first 10 years, my job was to anticipate how other people are doing and pattern recognize and put a
38:33trade, get in, get out, get in, get out, avoid the bad things. That's how we think. So and if you see 1988, where systematic trading came in Rentac, Evelyn Burlicamp was the first guy who really took technical signals, short-term technical settlement. wrapped them into quantitative units, then under statistical significance deployed them. And he basically did it in commodities and effects. Equities didn't come there. So there is a big debate in our community where you really go into the nitty-gritties of pattern recognition. There's a big overlap between technical and quantitative. Now, is that systematic? That's how. systematic trading started, I would say. But systematic trading needs execution to be 100% machine driven. So I never came from that school. So my my brain, my my my vision and all was to create something which is extremely rule-based and disciplined, Cleaned, but where there is a human element of managing the book on that. So that's where I would put semi-systematic space,
39:43wherein 80 to 90% of the trades that we do are extremely machine-driven signals that we do. And that machine is a confluence of technical, quantitative to the biggest extent, and fundamentals to a certain extent. But the executions are different. done by human under certain rules. Like we can go back to a very good story here. Richard Dennis in 1980 started rule-based trading. He was the guy who made most money in Chicago paid from trend following. So he gave rules and told, I'm going to hire people who have nothing in finance and I'll give them these rules. and they will hand execute. And he called them turtle traders. And those turtle traders just became really good trend followers till 1987. Richard was crushed in 1987, but then some of them lived after that. So would you call that semi-systematic? Probably at that time when personal computing wasn't to that extent, I would call that a
40:50semi as systematic. semi-systematic. In today's world, where our company, FR, is extremely quantitative, adjectivic, signal driven, and a lot of research driven. But at the end of the day, the trading is on the executions, guys. Every trade that goes into the book has to be vetted and pushed by a human. And that human has a rule that when all these things have a confluence score of this and that, put the trade. So that's why FRGM Capital, we have uh we have 23-hour trading all across. So there's a full coverage because I wanted to keep the day trading culture because I come from the day trading culture. I remember 2012 and all, I used to trade oil, and me and my friend we used to be in. Mumbai. So we used to have 12 hour shifts. Like when we take out the book, we give the book to the guys in London. So they run the book for 12 hours. Then we get the handshake book and all. I always loved that where humans are always the f last eye. Because Rahan, I'll just go back to the same thing. Like machines can do it, but when humans go crazy, someone
42:01has to manage it, right? So you just fall back into that bias that my brain always has: that humans can go monkey. So I would never want FRGM to go into a fully systematic hedge fund. So that's why we are semi-systematic, and that's how our style is, which is we are 90% 20% machine-driven machine driven signals, but the executions has to go through the green. light of execution traders who are covering 24-hour market. That's how we are semi-systematic.
42:35Unidentified speaker (speaker_1): That makes complete sense. And you know, since you've spoken so much about global macron, you've had experience working as a market maker, as a discretionary portfolio manager, and you've done day trading as well. Like, what do you think each stage has taught you? And how do you think that unique perspective is like sort of helping you get an edge over the other funds out in the US?
42:58Unidentified speaker (speaker_2): Yeah, the first rule is be modest and humble. I would not have edge over smarter guys. There are way more smarter people than me in this world. My job is to not do mistakes. And as long as I don't do mistakes, I will be on the uh other side of the story. But yeah, to your point, I've been doing it for 18 years. Uh I've done day trading initially, we were pretty successful, and then we had a blowout in 2012 when I had to go back into sell side because I got broke completely in 2012. Then I ran with a cell sell side as a quant research, market making. learned a lot throughout that time. Algorithmically worked with some of the smartest quants in the world. Learned really a lot from how different flows work, especially in market making space. Got a relationship with some of the biggest portfolio managers while I was a market maker. Really
43:59understood how they think, then had the opportunity to. money manage with two to three of the biggest portfolio managers and really learned how less I know. And over time I understood that through all these spaces where I was able to understand everything that I don't know, I want to start a fund in which I will never do those mistakes. So that's how I started FR three years back. Initially, I started it keeping Ray Dalio's philosophy that at the end of the day, money making is monetizing research. So first build a research company. Build a research company when you have an edge. And that edge can be so micro that it becomes noise in front of the world. But for you guys, it's an edge because you you can look into the DNA of the microstructure of that market and get getting in an edge. And if you see the history. every fund which has grown, they all come from one small durable edge. Like what was Edward Thorpe? Convertible Warrants, right?
45:03Citadel, option pricing. Saskena, Jane Street, ADRs. And Peer's Trading, Morgan Stanley, D Shaw came from very similar edge. Peerce Trading, because D Shaw was a part of that same Morgan Stanley group. Rentec, I would say people say it's a black box, but the Rentec inflection, because Rentec was losing money, Monometrics was losing money till 1988. 1988 Medallion had the year when they made 80% profit. And that was the year when. where they where Evelyn Burlicam came. And Burlicam really built technical screeners and wrapped them. So it was not rocket science in 1988 how they started. The first thing is Rentec is such a closed book, people don't understand. Gregory Zuckerman did an amazing work. I've been studying Rentec for years. If you really simplify everything, if you single
46:03decompose every one of these black box strategies and discretionary strategies and really go into that, you will always find one thing is the edge is so simple but so micro and repetitive. And your job is to anchor to that. Agnostic of which market, agnostic of whatever is happening. If you can parametize everything in the world. but still be married to your one edge, there is no way you can fail. Because no one knows what's going to happen. The only thing you have to do is that in the law of large numbers, your edge should work. So that edge can be gamma scalping, that edge can be Elliott way of pattern Patent recognition, that c edge can be an agentic way of finding some alpha somewhere, or that. edge can be something that Easy Englander thought that okay, I would just make the entire world a short gamma in front of me where I have some money, I will go and hire the smartest of the people. Whoever loses, I'll cut them off, and whoever is increasing, I'll push them. So Millennium came with this innovative way of really making money out of amazing risk management. Easy never came. with his own edge in 1989. He came and hired some of the
47:19smartest people. And he has this rule that you get a book, I'm giving you money, you go and hire from the street. I don't care what you do. It is your pod. The moment you're 50% down, I'll cut your risk to half. The moment you're 7% down, you're out. And imagine 36 years, one year they had a drawdown. Every other year they worked. Well, easy and done through a lot. Easy went through insider trading scandal in 1994 and all, but those are different things. I would not go into that. But Millennium, Millennium's edge was trying to synthetically create a short gamma out of PMs. That's an edge. You just need to know where on the micro you have the edge. You have to keep sneaking for that and you have to do it because I would say at the end of the day you should always defocus and see. We're just talking about trying to get an edge in 40 years of evolution out of 2000 years. So none of this edge are sustainable. But you just need to know where there is statistical edge, law of large number in a microspace which is not decaying and something which you can find in economy of scale and scope. And your your aim is every money that you make is an inventory to collate and keep repeating that edge. Everything else should be taken care of. So, worst
48:38thing is you have an edge, you make some good money, and you forget that the money that you have made is because of that simple edge, and then you try
48:47Unidentified speaker (UNKNOWN): to
48:47Unidentified speaker (speaker_2): do a lot of things together. A lot of people. have blown out on that. So our philosophy is like we have, yes, we have secret sauce, we have IPs which we protect, which are our edge, and we keep snipping into the markets to see where those edges are. And we keep constantly repeating that durable edge across different asset classes. But that edge is not like a money-making machine. That edge needs constantly to be better in technical charting, better in quantitative analysis, better in fundamental thinking. Everywhere you need to be marginally better every day. So we have a quantitative side of the pod, we have a fundamental pod, and we have a technical pod, part, and everyone is challenged every day to marginal. marginally improve. So probably the edge is to restore the edge and constantly improve, I would say. Otherwise,
49:49you know, you can just come swing out and come back. I mean, those guys also work. Like Arini, if you see this year, they are down right now 20%, but they swing, but sometimes they have huge. asymmetric winside. So it all depends on how you take your style, you know.
50:08Unidentified speaker (speaker_3): So you spoke about the challenges that you faced in 2012. So if I have to ask you about your learnings, about your edge, and you know, to also move forward, how do you describe your edge now and how has it evolved over your years of trading?
50:22Unidentified speaker (speaker_2): Well, I'll tell you basically we had a lot of positions in spreads and flies and then and we used to think that PC two, PC three are pretty protected. They would not move. But then oil market just moved and we just saw that we got completely hammered. I realized that quantitatively we would have needed more scenario analysis and more higher dimension hedging and all. So yeah, technically you learn a lot of things. You get it into your book and try to live. But what I really understood more was risk management, I would say, and not get super excited by being little successful for a few years. What that means is like there are two to three ways in which you can take this business. I think the success story comes when you when the need for you is to make sure that your business survives for hundred years. That is your only aim. Money, money is a byproduct. Money will come and go. You will
51:25lose sometimes, you will make some time. Don't go on a victory lap or don't get depressed with the macro mayhems of the world, but try to be so. disciplined and structured so that you can live for the next hundred years. So I would say that's the biggest learning that I did because uh we couldn't live. Like I was running some money with a very small hedge fund, Oak Ray Capital and Hall, but we just couldn't survive in 2012. And uh that's where I moved out of commodities and energies into into rates. I mean, at the end of the day, anything moving in the world, if you parametize stories become very similar, you might understand the intricacies of one market versus the other. Yeah. So the learning comes is I think the same learning Soros had when he lost 300 million in 1987. Now, I mean,
52:25you just don't know, but it has to do a lot with what kind of personality as a CIO you are. So I realized myself way more as I kept I had so many losing stories over the last 18 years. It will take five hours, so I would not go into them. But every time something like that, what Rehan, you're talking about 2012, something like that happens, I would have a small notebook and I will note it down that my job is. to never repeat that mistake again. Let me do other mistakes, but not put everything in spreads and flies. So if it moves, we are done. No, no, no, no, no. Don't do that. PC2, PC3 doesn't hedge you everywhere, you know.
53:04Unidentified speaker (speaker_1): I know you sort of touched upon this earlier where you spoke about how AI is making it easier to generate research and like develop trading signals.
53:12Unidentified speaker (speaker_2): But then whether that research is you useful or not, my my point, Krishna, was a lot of repetitive task is taken care of with very less margin of error by agents. That's all I'm saying. There's nothing more than that. I would not find a great CMT who can count Elliott Wave by trying to train an agent. I had several discussions with my CTO and others. Like my firm is crazy about AI because all these are CS guys around me and they are always like, no, we will revolutionize everything that we can do can be replaced. No, no, no, no, no, no. Like somewhere Stan Rakamilla is a great chartist. I just don't think agents can replicate Stan's kind of chart reading, Paul Tudor Jones kind of chart reading. No, no, no. Because because of the factor that there is fear and greed. and there is sometimes when idiosyncratic stuff happens, and supervised learning cannot capture that. So no agents, after being trained, whatever way, cannot capture that. So research is getting an edge, Chris, for sure. The same thing happening In in consulting space, I would say, or in a law firm, I would say, like paralegal's work might. have mistakes and all. Paralegals might have taken more time to write something.
54:32Now it gets to you. The same thing, like in our space, what we do is we always compare apples to oranges, right? Like election plus one, how does egoodies move, right? Before in our days, we'll go into our data, do a time series and all election T minus one. What are the midterm elections when it moves? That would need one or two analysts. Right now, a Claudia, you can just do it very fast, and the margin of error might be less. That's that's a paradigm shift in research. I believe whichever style you are, discretionary macro, which if you are fundamental style or technical style, quantitative style, in our human brain, all we do is we pattern recognize. from whatever has happened historically, right? Now that we have come with agents who don't get tired and you can they can do certain works without a lot of margin of error. Because before, like machine learning, this is not the first time when machine learning came. Bamwell's algorithm came in way back in the 1960. I think the change that has come with AI is the margin of error has gone down. Like I go got really surprised in the last two to three years, like agentique-driven analysis are sometimes so close or better than human-driven analysis. I think that's the part which research is getting a
55:52huge edge. Like I can be a one guy startup, but I can still really run a couple of back tests and do certain stuff and give information useful, like like web creators. Right now for you. to create a website, there is there's no concept. You know, you can just do it like that. So I think those spaces are the places where research got an edge. And the idea is as humans, we should know that our job is to use it on our side, make it like a catalyst to make FR Global Macro to next level. So I'm pretty excited and my entire research team is very excited. We massively use AI all across. Our cursor and clock costs are out of the roof, but I love it. But yeah, so research got a huge edge, but trading I don't think will ever have an edge.
56:40Unidentified speaker (speaker_1): But do you think just 'cause now everyone has access to the same sort of research and as these models keep on getting more advanced, like do you think the life of the investment or the edge is reducing?
56:52Unidentified speaker (speaker_2): No, it just goes back to the same concept of Eugene Fama and Paul Samuelson, efficient market theory. When material public information has access to everyone, does it reduce down the edge? No. Because then the edge comes in a different dimension somewhere else. The second order of that implications. Personal computer was not there. D Shaw, Donaldson, Donald Sassman, spent $2 million in giving two Ferraris to David Shaw to start 1988 to 1995. Deshaw has a great age. Today, does that the age for D Shaw probably not? A very interesting story. Like previously, for Rentec, Brentec, Millennium, and other places, some of the biggest IP for them was their data warehouse, right? Now, do you think that will be the same story in 10 years? No. Data became so much like data is in front of every one of us and it's getting fine-tuned and fine-tuned. Like
57:52if you go to old school, the stories like they used to have ticket tracks. Some of the work of these people will Where just come and note down the data and push it into your ticker ticket trade to make patterns and all. But right now, everything got disintegrated in a place where you can have like I think it's very easy to get tick data right now for anything like price versus fundamental tick data. You can get everything everywhere. So, what does that mean? That basically means like edges will be explored in some other dimension. That has been the story. Like, AI will not make sure. that okay, from now onwards, everyone, socialism comes into the world, everyone makes the same amount of money. No. Like the Austrian theory of economics always works. Someone will get penalized, someone will go, mistakes will come. But then people will think on the next level, people will think on the next level. It's not stupid that Citadel is hiring so many guys from open OpenAI AI who have no understanding of finance. sitting them there and trying to think like my management, we our biggest focus is to hire the most AI-centric people in the world. I mean, that's not because we know that we are getting wired. They are like the next story. No. That's because like every time humans need to integrate different thoughts and edges, and
59:11another dimension of durable edge comes. So we need to know that even everyone knows the data. What is the other thing that can happen outside of that? It's like game theory optimal, like there is always an edge on the next step. That's why we are humans. That's why me and you, after 100 years, will talk the same story. Even if we are in Mars or somewhere, there will be an edge somewhere. By the way, do you know Harappa civilization has three layers and Vijayanagar in India? people used to sell diamonds on the road. So maybe 1000 years back there were spaceships and all. So we are a little more myopic to think, oh, this is the place from where everything will be standardized. No, you just never know. Maybe this evolution goes for 50 years, paradigm shift, everything starts back from crash. AI is definitely going to change things in Wall Street, in Bay Area, in other places. investments, and you can see it everywhere in MA spaces as well, as you see. But I think this is way more myopic and tactical to think. Like things will humans always figure out a way to anticipate what the other person thinks is happening, and everything is standard, and play little more than that. That's where humans do. We always go to the next level.
1:00:25Unidentified speaker (speaker_1): In your opinion, where do you see? the industry progressing as a whole? What do you think is the next immediate outcome or second or third order effects of AI?
1:00:32Unidentified speaker (speaker_2): Industry as in you're talking about investment. Okay. I would give it two different factors, right? One is say a firm like Goldman Sachs, they are integrating AI to a great level, but they have to do it way different. differently because they have an already huge infrastructure. They have a pillar where they make everyone AI educated before deploying it. And then there's a point where they would deploy it. Versus you have someone like Perplexity or a Uh budding hedge fund like FRGM, bunch of smart guys coming, everyone knows certain things, don't understand the other side of it, recreating stuff and all. So two different schools. Some of them they have to preserve what they have. So there's a very different way in which you will absorb this new world versus it becomes so much competitive for startups and other places where
1:01:35you can really play it faster. So both of them have potential bad scenarios that can play out. One is when you're taking too much time, you can miss and you can be behind the curve in a bigger organization because your research might not be as good as another's research, which are very agentic as
1:01:58Unidentified speaker (UNKNOWN): well.
1:01:59Unidentified speaker (speaker_2): And versus if it's not robust,
1:02:02Unidentified speaker (UNKNOWN): if
1:02:02Unidentified speaker (speaker_2): it's uh you don't know the contrarian part of it that if it doesn't work what happens back so that part in our world what we call version controlling I think for new companies which are coming out they should be very focused on that part that's the point everyone is talking about there needs to be more control into the picture so but then I would say both of them are realistic because if you're a startup, if if you're getting such an exponential edge of using agents over time and then can compete with the big players, they would definitely do that. You won't be able to stop them. You can put a huge lot of control. Well, US haven't done that under the Reds, which I understand. But how much it will be great on or not, that's still a debate. I don't know. Like what control where you want to do, how much you want to do, where you want to do, how smoothly it comes into play, I don't know. And it all depends on industry to industry, in my opinion. And of course, government policies, and let's see where it ends up. I don't know. I don't want to commend, comment, I don't have a bias into that. All I know is like tactically it's inflationary. at this point. The growth story has been pretty good. Like in month of February, you remember when the
1:03:21Citriani article came, there was this huge rust of disinflation and everything going into like I can deja vu another scenario like that coming, maybe now that equities are going into weakness and all. Europe is very skeptical. They have always been. That's the culture. They're way more apprehensive and all. Versus, if you see in Asia, especially in Japan and all, even their MA activities and flows have really come up more, if you really see. So, like China, domestic part, they are not doing that much, but overall, in innovative AI, they they're are doing really good. So overall, globally, I would give it both the factors. Like big firms, they are taking it a little more timing, but they are coming into it, which is fine. Control is a very important factor. And now I think in US also it will definitely come up given the elections. But there are two sides of the story, I would say. Agnostic of what happens in the market if you real
1:04:25Unidentified speaker (UNKNOWN): if
1:04:25Unidentified speaker (speaker_2): you think about the real growth. So nominal growth has been six percent, real growth probably around four, but six percent is pretty decent in my work. MA activities have been pretty good in the last two quarters as well. So I don't know what happens after this week. Things can change. So I would just say these two pillars should work around each other. Like, but it's definitely an edge for small. businesses, for startups, for firms like research companies and hedge funds like us to compete and they would jump into it.
1:04:59Unidentified speaker (speaker_3): Yeah, that's I I totally agree with you
1:05:01Unidentified speaker (UNKNOWN): on that.
1:05:02Unidentified speaker (speaker_3): Being new and lean does have its benefits. So so talking more about FR. So how do FR Global Micro and FRGM capital complement each other? What does one give the other that it could not get alone?
1:05:15Unidentified speaker (speaker_2): So it just goes back to uh that point which I really love about Ken Griffin. He always says the pillar of Citadel is Citadel was one of the first firms who were able to monetize research. And maybe, I mean, he has done phenomenal through GFI Cityl Security. I believe I'll do better than him, but let's see what happens. Uh
1:05:37Unidentified speaker (UNKNOWN): one
1:05:38Unidentified speaker (speaker_2): of the reasons. we started with FR Global Macro being the being the research hub which continuously grows, and FRGM Capital being the hedge fund which is come which is way more connected, gets exclusive access to all the researches. And but there's a very strong Chinese world because we have other client relationships. but it becomes very symbiotic. I'll tell you how. As a hedge fund, because I'm the CIO of FRGM Capital, so as a hedge fund, you are always matted to your book, to your positions. Now, research should give you a spot information edge all the time. That is the job of an ideal, godly research kind. of an environment in which a PM can sit. Let's say I'm a PM, I'm sitting somewhere. If I have five God hands, a single bomb happens somewhere in geopolitics, like something happens. We are the first to get the information. We are first to see what the geo-score is on that. We are first to see how apples to oranges you you can can compare whatever happened right now to the last five instances. and what has
1:06:51happened on the market on that. We are first to see what happens T minus 1, T minus 2. I mean T plus 1, T plus 2, T plus 3 on that. We are the first to understand the contagion effect. We are first to see that what positioning can change. That's what research is, right? Research is like when something happens, make me like a god. I have all the information of the world before I put a trade. or before I take out. Ideally, that is the symbiotic part because FR Global Macro is always hungry and asking the research the right question to do. And the research company in itself, which takes care of FR versus takes care of other companies as a research provider, its aim is to always grow and cater. So for FR Global Macro. FRGM Capital is one of the important clients. And our style as a semi-systematic global macro unit is somehow shaping a marginal quantitative edge in the research that FR Global Macro gives. But FR Global Macro, as I spoke, has three pillars. We are one of the pioneers in Elliot Elliott Wave. because I personally believe I'm one of the best Elliott Wave counters in US, in US Treasury. I mean, it comes
1:08:10from a lot of history, and I'm just saying because it's a fact, in my opinion. One of my partners in India, he's one of the best Elliott Wave technical guys. So I think definitely we have an edge. Like David Rosenberg has an edge because of his. fundamental aspects. Pantheon has an edge for Samuel. So when we started three years back, if our global macro, technical and elliot wave, we had an extreme edge because of the years of experience, so much of fractal analysis that we have done in technicals and all, with my years and some of my partners' years and all. And we are constantly. hiring CMTs, training them over years across different asset classes. So, yeah, so that was our edge. So the first idea is to preserve that edge over years because of my bias. Like 500 years back, technical analysis is always having a great way to understand human emotions of fear and greed. And into the micros, it helps the portfolio
1:09:17in so many ways, like gamma, price discoveries, strike discoveries, a lot of ways, a lot of things which standard uh firms don't look into. You can really use technicals to make the portfolio management way more efficient. At the same time, we are extremely competitive on the quantitative and fundamental side, and we are growing a lot on that. And that comes from this symbiotic relationship because we are a semi-systematic fund. And so there is always constant, huge push of going to the next level, do it a little more marginally better. So so that way our quantitative signals are also. extremely good. When I'm saying our I'm talking about FR Global Macro, and we have other institutional partners who collaborate to a very strong extent, and we are overlapping, and that's growing a lot. But the relationship of FRGM Capital and FR Global Macro is so symbiotic that it helps both sides of the unit. But these are very completely separate units. There is a very strong Chinese wall between them, but they're growing very good together.
1:10:31Unidentified speaker (speaker_1): What do you think are some of the challenges that you face while building both the research house and the fund together?
1:10:37Unidentified speaker (speaker_2): Well, not I would not say because I'll tell you a story. Uh Kobe Bryant was asked why would he wake up every day, four o'clock in the morning? The entire concept of black mamba mama mentality. and all? So
1:10:52Unidentified speaker (UNKNOWN): one
1:10:52Unidentified speaker (speaker_2): of the things that I loved about him is age 13 to age 17, right? He would just dribble the ball to sleep. He would do everything. And then people were saying, as a kid, why would you do it? So his answer was like, to some kids it's broccoli. To me, this is French fries. So the answer is the same for me. So firstly, there is no technical challenge. We have two separate legal general counsel, everything is well. Well, as a founder of both, does it make my life challenging? I mean, I couldn't live another life. This is I just love it. See, structurally, there was a very strong thought that why you want to grow something which has so much of synergy, a research house which gets collaboration with other funds. other they might be your competitors, they might be self-side banks and all, but there is a constant information that's coming towards versus there is a fund which is growing, which is trying to use that research
1:11:56as a place where you preserve your IPs, but you keep growing and you keep testing. all the good things that you're doing versus all the wrong things that can happen and you can mitigate. But of course, there are places in which you see in the past there is a big legal challenge.
1:12:17Unidentified speaker (UNKNOWN): So
1:12:17Unidentified speaker (speaker_2): we gotta be extremely careful on conflict of interest and all. So we are very focused and very legally. disciplined and responsible. But that's not a challenge. That's that's how you should be. That's like a that's like a rule in life. I don't cheat, and I'm honest. So that's the same thing here. Where if you really talk about any other challenge that I can face, I mean I really don't care about financial challenge. I really don't care. about uh market volatility as a challenge. I just love the fact that I get this opportunity to interact with some of the smartest brains of the world and give them this entrepreneurial dream. And the fund has grown to more than 20 to 30 employees by now. And every guy I might have probably rejected twenty 20 candidates before we picked, and we just keep. growing and growing and I couldn't do anything else other than this. So I'm extremely grateful.
1:13:21And I am I'm a student of history, so I can just think I'm like Ken Griffin in nineteen ninety three. That's all I think. Or I'm like Easy Englander in nineteen ninety one. Same thing. By the way, uh Easy was forty forty-one one years, nineteen eighty eighty-nine, nine when he started Millennium. I'm forty forty-one one. now. But I started three years back. So you never know.
1:13:42Unidentified speaker (speaker_1): Yeah, you never know. Yeah.
1:13:43Unidentified speaker (UNKNOWN): Like
1:13:43Unidentified speaker (speaker_1): I'm anything's possible in this day and age as well. Um, I guess one thing that I wanted to know is how do you decide what to use internally for the fund versus what you share as a research product?
1:13:55Unidentified speaker (speaker_2): Oh, whatever you're sharing with the research product is not under my purview. Uh so we have a chief strategy officer, and that's a different manager. management. So I am the active CIO of the FRGM Capital. And I interact with my peers and clients and all as just PMs interact. That's it. So whatever the research share is completely by the active management of the research. That is not what I see. They have their own chief strategy officers, they have their own head of research, so they're doing that. I get what I get and I put my trading based on that. I'm always focused on my book.
1:14:32Unidentified speaker (speaker_1): When I read many equity reports, like I'm always wondering, should institutional research give like explicit trade ideas or should their main purpose just be to educate the investors?
1:14:44Unidentified speaker (speaker_2): Let's first understand. You give a trade idea, it works it good. If it doesn't work, it's still like okay. So what is the difference? between giving a trade idea and giving your own bias? I don't think I don't think there is not a huge difference. So you can structure that given this is the bias, I think this has an asymmetric payoff or something. This is how I want to structure it. That's your own thought that you're sharing, versus Is on every relationship, even even after you give City 7, City 63, C7, C seven, like there is. always a thing that when you're giving your trade ideas, it is on the full discretion of the recipient to take it and use and whichever happens in that way. So now comes a question, are you getting paid for that? Yeah, but so that's that's very different research where cell sites have it, where like technically. you're being paid for giving trade idea. That is kind of an advisory research. Versus market intelligence research is like you're just showing what the market is doing. So you're just giving a service and getting paid. That's what I think FR, FR Global Macro is a market intelligence research. It doesn't give trade ideas to anyone.
1:15:57It just shows what's happening in the market, what he it it as a research unit thinks about it. Think about it. If you are if you are an advisory, that's a very different role, right? So I don't want to comment on what should be right or wrong to be a advisory. I don't know, but most of the institutional research units are more focused on giving market intelligence and what they think rather than really giving trade ideas. Yes, banks do give. There are different ways in which you get ideas, but then uh PMs do their own stuff. Like, okay, I get ten trade ideas from ten different research units and banks. FR is one of them, though FR doesn't give, but other ones give. I look into them. Sometimes Sometime you might use, sometimes sometime you might not use. I mean, that is your decision, but then
1:16:50Unidentified speaker (speaker_3): Yeah, that's fair. So moving on to the current scenario with the Fed navigating persistent inflation, geopolitical uncertainty, and changing growth expectations. How do you interpret the current US rates environment? Where do you think market expectations could be wrong?
1:17:08Unidentified speaker (speaker_2): Well, let's just see what happened through the year. I think that is a that is a better way to understand where. we are. And if you're talking about rates, let's just see through that. Let's look into tenure. Like, okay, January was a big corporate issuance, and market was, I mean, rates were like bonds were selling off. And then the month of February, there was a blockbuster NFP, 5th of February as alone, well, but the market just faded it. So somehow the bonds started. rallying. Month of February. Bond started rallying. CPI and all, the PC deflator and all were still showing inflationary. Mid of February bonds started rallying. Now why? Because the AI fear started coming and Dan Lobes and all were really hitting the tech sector that time. I'm talking about mid of February. Okay. Uh Citrian Citriani article came on. disinflation and all, and tenure went to 385. Now, end of February, Trump hits the Supreme Leader. And that Sunday was the top of those bonds at 385. And then you go from 385 to 535, right? 140 base basis months, 150-based 150 basis months overlook. over like So, what is this story of 150 base? basis months? Now
1:18:28from March to now, what has happened if we really see not just Fed, what has happened overall in macro? You got into this war, which is so getting into the war month of uh March and April, people would probably think of it since the petrol petrodollar dollar system, every time when there was a war, US trade. Treasuries have always rallied as safe haven and all geopolitics, Middle East has always bid US Treasuries. How from the inception it is taking it so inflationary and bonds are getting hit, bonds are getting hit. I mean, I think uh Jeffrey Gunlock and all were absolutely right on the 30-year and all what happened, but I would really categorize it more into a unique. situation of extreme geopolitics with AI playing like a binary unit. It can be the Capex Saga versus it can destroy job and disinflation can come. And it played the Capex saga through that.
1:19:34And whatever the state structure was with whatever. Trump's approach was into that. And speculators getting destroyed. Extreme positional bias into that. Now, is it like the first time? No, of course not. These kind of these kind of pain stretches in rates market has happened before. Now, a few things. Let's count how many how many global micro hedge funds are blow
1:20:02Unidentified speaker (UNKNOWN): up
1:20:02Unidentified speaker (speaker_2): through that. Tola blew up, Cyril was getting crushed initially. Most of the funds on the macro space, like we have seen so many evidence of a hawkish print in tier one data, and intraday market is completely fading on the other side, giving cross-asset beta an and edge. In my 18-19 years, this doesn't happen that many times. So there has been huge volleyball in our space. And that has really made this market very tough to navigate. Now, I would not say it's an inflation story or it's a hyperinflation story or it's a stagflation story or it's is a disinflation story eventually. Uh yeah, okay. Now when with war coming, there is a diffusion matrix in which we are seeing forward inflation story. Well, I would look at it more like there is more uncertainty in the way the last six months have been played. One of the primary reasons for that is this is like this unique combination of
1:21:12a political strategic. view, which is little more on the stretch side, and at the same time, a very uh painful positioning in our markets. And that's where rates got hammered so much. Like that's where we see so many unwinds wines in macro pods in rate space. Now, let's categorize during this. the six months how does RVs would have performed, how does discretionary macro overall would have performed, how does systematic funds would have performed, versus in equity, what has happened. So I would say month of February March, RVs would have got killed, which has happened, uh like starting from most of the places, including Bluecrest and all. Then if you see systematic should have performed more consistently once the wall-of-wall adjustment came by month of let's say April, May. Discretionary things have moved idiosyncratically to a crazy extent. Some of the bond faders, like Jeffrey Gunlock, were talking about 30's 30 stealing tailing for almost two
1:22:23years and now he catch catched the train. I mean, could kudos to him. But I mean, what was the thought when we were at 385 in February, right? It's not so it's very different than that. But my point is this volatility is not going to go away because positioning got tested, and PLs have been pathetic in global macrospace right now. And there was always this, there was always this confidence of big players who can run through painful stretches, but can mean the word bar, like deja vu of SVB, where Rokos went down and still ended the year up. Well, that didn't happen this time. People stopped out this time. So market positioning is lighter right now. So trend will come more into our space. Now, well, December is 16 base price. Like, I don't know what the any like what the tactical uh Fed policies are gonna move. I never bet on them. Like, of course, we have views. I can go into details of that, but I don't think that will really be great for this discussion. But on the micros, markets
1:23:38will move. over stretch, tier one data to data, stretch to stretch. On the macro, I do know there is an affinity and love for selling long and bonds. Uh, I just don't know how good that will work if the equity sell-off keeps going more. Now, This CapEx hit is probably going to. change the sentiment. Now, if you go out of all these things and think more from real money perspective, it's a complete different story there because the price insensitive real money part, I there has been two times where they try to enter, and I just don't think Thing in rates there is that much of love of buying 30-year bond till the time. it's 6%. So we went to 575, but then also real money interest wasn't there. Versus if someone tells me, can we just rally back to that Citrene article kind of situation? The answer is absolutely possible. Why not? Because if equities really get back to a risk-off mode on that, that was the normal.
1:24:51after that all this happened. Now, now let's see what kind of relationship this has to two of the most important factors. One is the war, as everyone is talking about, and the other one is AI capex. If you think about from the war point of view, the risk premia in our space, uh, it is very heteroscritic. heteroschedastic at this point. It's like smaller risk premia from sudden time, boom, it's increasing. The moves are coming higher, like de-escalation and escalation have higher risk premia. Again, it's going into smaller, smaller, boom, it's going up. Now, there are times like Friday's NFP, gamma got hammered, like it didn't deliver anything, versus there are times when a simple service PMI is delivering 5x break-event. So I would say markets
1:25:46Unidentified speaker (UNKNOWN): are
1:25:46Unidentified speaker (speaker_2): very stretched in our space. And the concept of durable edge that we talk about, that has gone too. Like that's not there for discretionary macro, I would say. And to be honest, like you gotta be way more longer handed. horizon to really think of a durable edge like that, especially when you're dealing with something like macro. Macro can be idiosyncratic and macro can be extremely non-fundamental for a long period of time because there is so much of positioning and other things that deal with it. Because you have to understand 70 to 80 percent of the flow of macro backers is real money, which are priced price incentives. insensitive. A lot of these, some of these biggest life insurance and pension funds of the world, a lot of them doesn't have a huge incentive of making a lot of money on the prop side. There is no concept. So, but the 20 other percent of it, which is the speculation and the fast money on us, the effect of that sometimes gets extremely
1:26:48strong. And we just go into this zone. that we are like, and how is the is the world crazy? How how is I don't know, like how is 30 years of the 575 and standing? Like, of course it can be. It can be more. I don't know. Can we go back to Volker era? Probably not. There is no it's it's nonsense, doesn't make any sense in my in my opinion, but that's my opinion. But I would just give it a little more. parallel comparison. So imagine commodity corporation guys in the 1980s trading oil, cotton, sugar. Well, is there idiosyncrasy? Can the market be overstretched on them? Of course, they went through some of the biggest trends following time. Now, rates was not like that because rates is very fundamental and everything needs to play into place. But I just explained in the last five minutes all the factors that can move us into a space like that, which happened, right? 150 basis point moves in rates through a through the war time where hardly any mean reversion. Well, okay, you have to understand this is the first time in which a Middle East war is really. selling off bonds to that extent.
1:28:05Geopolitical spike has generally been
1:28:08Unidentified speaker (UNKNOWN): a
1:28:08Unidentified speaker (speaker_2): bond rally, a flight to safety over years through petrol systems. So that's kind of the answer that we have to be a little more cautious of how overstretches can be handled in global macro Micro for the foreseeable future. And uh I think once once the war eventually gets out, then things might go back to normalcy, and all the micro edges will start working back, in my opinion. But it is where it is, and just see what happened in long short equity just in Thursday, Friday, like two days. But isn't that the same story that has happened all the time? Like, why we get surprised on the same thing that keeps happening over 40 years? Like, no, there's nothing to surprise. Like Howard Mark has been telling there's bubble, this bubble, this bubble, everything overpriced. But I don't know. Like, I won't be surprised if this sell-off keeps happening. I mean, I would rather be happy making money and not being right. So my my job is to just know
1:29:12Unidentified speaker (UNKNOWN): what can
1:29:12Unidentified speaker (speaker_2): go wrong. Like at least okay, sorry. I think that was not humble to say. I would rather be happy not losing money. And making money will take it back.
1:29:21Unidentified speaker (speaker_3): Coming towards the end of our rather very enlightening discussion. What do you like to tell students and early career investors who are trying to build a macro career?
1:29:31Unidentified speaker (speaker_2): Yeah. So I think this is something that I keep talking to uh people whom we constantly hire, we mentor. over time. Is the first part of the career is to understand what you're good at and what you're not. And though it's a very easy question, but the answer is very difficult. Uh let's take an example of, let's say, Michael Pelps, right? Michael Pelps Phelps had to build his career from the age of 13 to 25. Sydney Olympic 2015, he came fifth. Athens Olympic, he was uh 19 years old. Got crushed by Ian Thorpe, called the race of the century, but still got other places. But what I'm saying is every four years for a teenager, the marginal improvement that he had, which is not correlated to money, not correlated to status, not correlated to how people are appreciating you tactically,
1:30:32some some humans have to do it at the teens. I think in finance, this you have to so you have to be like an Olympiad from 20s to 30s, where you have to always forward think. Think about this generation. Let's say both of
1:30:48Unidentified speaker (UNKNOWN): you.
1:30:48Unidentified speaker (speaker_2): AI is going to change the world. to that extent. Of some of the world that you don't know, I know, but AI is going to change it, right? So think about it like you are Olympiads and some of the rules are getting changed. So what kind of training you need to do for four years to still try to get a medal? Firstly, understand what is your passion, what do you love, and you can integrate that into making a good living and money in in this space. So once you understand your passion and love, ours will become like a byproduct. You don't need to take like you would not get exhausted by putting something into the hours. And I would be saying that you guys are one of the most lucky to be born in such a time when AI is taking over. where the independence through which you can run your career has been so high to someone, let's say you're a developer 20 years before
1:31:48that, you might be not good at C and Java. It will be tough for you to get into a place like Red Tech on those days, Microsoft, Google, or even to let's say Citadel. While Why because the syntactic sugar. is so much related to how good or logically or algorithmically you're thinking or not. But that's not the case right now. You know how much it changes people's career right now? Now the importance is how you can innovatively think the next level, how you can recreate Create things. So so what that means is the definition of success for your generation is pretty much different from what the last 30 years of generation of general schooling and success has been. And so try to understand what you're good at, what you're not good at. Try to identify what is that one thing that you want to do and live and die. Like for me, at the age of 19-20, nineteen twenty, I realize it's trading. Trading, post poker, I was just I wanted to play with probability and uncertainty. That's what I always and I had a very I knew I had a very stable nerve. I never get scared in a bad situation and I understood that was my strength. Look, twenty years I'm
1:33:07I'm still fighting and probably I whatever times I will live more, I will try to do it. So that's my suggestion. Like try to find passion. And when you when you love to do something, if you're passionate about something, the world will conspire. The world will somehow make sure that success becomes a buy it becomes a byproduct, you know. Think yourself and with AI and all, right now, I would say the barriers to entry has become soulless and it's going to marginally get lower and lower. So it's like an even race. So gear up and do your stuff there. And I would not define it whether you want to go into Wall Street, you want to be a buy side, sell side guy in our space, or you want to start your tech startup company out of Silicon Valley, or you want to be, I don't know, a social entrepreneur. So, but I'm saying that
1:34:01Unidentified speaker (UNKNOWN): the
1:34:01Unidentified speaker (speaker_2): barriers to entry. have reduced. So it's way more important for you to know that you are in that four-year gap of one one Olympic to another. You are the Michael Pelps who just got out of Athens Olympics and is going into Beijing Olympics and wants to break Mark Spitz's eight out of eight gold medal record. So Michael Spilz Pelps uh won from the age 19 to 23. from Athens Olympics to Beijing Olympics at one aim. I'm going to break Mark Spade's record of eight gold. And in his interviews, when he talks about it, he's like, I'm con I am competing with myself. All I'm doing is sometimes he would deliberately swim and train with broken goggles. He would he would count his breeds. And all this time we were talking about edge, right? The edge of success is you can go so micro into your passion that you can see certain things and can win over those things which other people can't. I don't know whether you guys know cricket, which is very important. I think you guys know great. Sachin. Suchin, The only thing
1:35:11people used to say, Sachin Suchin used to see the ball like a sock. soccer ball. Now people used to not understand that how and why. I think it has to do more with the passion and the brain. Like the brain is a muscle. If you can go to that level, you can see the DNA and the intricacies of things, you will find your edge. And then it will never be a broccoli. It will be always chicken wings and French fries for you. So and with AI, I mean, this is your game. Like it doesn't matter. Like Peter Till started clarine broke out, right? He was a tech guy, made so much money, tried to be a hedge fund manager. He was hugely unsuccessful. Came back. Look what he's doing. So you can be the PayPal mafias. I don't know. And imagine all of them were not from Harvard's. Harvard, Stanford, MIT. Paul Tudor Jones was from University of Virginia. Jeff Yes, Reeves College. I want to see you guys in four years. You're like, okay, we prepare for the Olympics, like what Michael Pelps did from Athens to Beijing, I think, because this is the time. Everyone is at the same platform. So should run it.
1:36:23Unidentified speaker (speaker_3): Right. So thank you for joining us today.