Unlocking Venture Growth Equity in AI: Al Tarar and Rizwan Muhammad of Quartus Capital Partners

Andres Sandate: 00:00:00

I am excited to be joined today by the team behind Cordis Capital Partners, which is a venture growth equity investment firm I was introduced to recently. So joining us today are Al Tarar and Rizwan Muhammad. Welcome gentlemen to ATLS Andres.

Al Tarar: 00:00:17

Thank you very much. We look forward to having a broad ranging conversation about AI and other things of mutual interest.

Andres Sandate: 00:00:27

Yeah, for sure.

I mean shout out to our mutual friend Mark who made our connection introduction back in the fall at the FactRight conference in Nashville where my partner Brian Cody and I attended. Our wealth advisory firm, Gramercy Park Wealth Advisors is, you know, is a full service financial services firm.

And one of the things that we're spending time on is know thinking about ways to expand our alternative investment and private markets offerings to our clients. And so the podcast is a really great way to educate frankly ourselves and and more importantly our clients.

So while we can't and, and certainly as we've talked about not giving investment advice on the show, and whenever people have questions about investments, you know, they should seek professional help. Whether it's a financial advisor, a lawyer, estate planner, cpa.

The point of the show is really to give perspective from leaders in and across the wealth management, investment management, asset management industry. So we're excited to have you both here today. Would love to get some background.

So Al, I'd love if you could tell us a little bit about yourself and how you came to work with Rizwan to form Cordis Capital Partners and tell us about the business.

Al Tarar: 00:01:51

Sounds good Andres. So thank you again for having us on this program. My name is Afzal Tawar, I go by Al. My background is intimately involved in Quartus DNA as well.

So it's a think about it, our firm DNA and our personal backgrounds are intertwined in a very significant way. So to give you a little bit of background, I studied AI in college in the mid-80s.

So for some of your audience they can calculate how old I might be from that. So the reason we're talking about it is we've been talking about AI and doing AI, working with AI for last 30 plus years.

And I studied some more AI in graduate school here in the US so that has been the underlying kind of foundation on which Quartus Capital Partners exist today. So after I finished my graduate school, I spent 25, 30 years in management consulting working for some of the biggest consulting shops in the U.S.

including Deloitte, IBM Business Consulting, former KPMG Consulting, that became Bearing Point. I actually led their Asia Financial services practice out of Shanghai and Hong Kong.

And then lastly, I was a partner at PricewaterhouseCoopers PwC in their consulting practice here in New York as well as in China and Hong Kong. So that academic background, consulting background gave me two ingredients to start the firm.

And Rizwan added another dimension which he will talk about in a second. So Quartus, since I have a microphone, I can tell you a little bit about Quartus as well. Quartus is a New York City based emerging manager.

We are a new firm and we are raising and investing in 100 million dollar fund to invest in mid stage venture capital, basically series C, series B types. And our primary strategy is AI based solutions to some of the most difficult problems we have in society such as healthcare security and so forth.

I'll stop there and pass the microphone to Rizwan so he can tell you about his background.

Rizwan Muhammad: 00:04:39

Thank you. Thanks Al. Yeah, so unlike Al, I didn't study AI in college.

I was a lot more focused on the hardcore, you know, engineering, computer systems engineering. That's what my master's was from Syracuse University.

And I, once I graduated I came to Bay Area and I've been here and you know, when you are living and working here in Bay Area, you end up with meeting all kind of technology people, get opportunity to work in great companies definitely. So I, I did work for some small companies, some large companies.

I was deeply focused on or interested in working for smaller companies because that's what excited me always. So I had about four or five startups.

I joined initially as an employee, but towards the end it was four people effort to launch a company which then ended up being acquired by Hitachi. So in terms of my background, it actually spans a lot.

I did software development, software engineering in the beginning, went into consulting, did that for a while and then product management. And that's where I spent about 20 years of my career. So that's my primary background. My AI encounter was at Hitachi.

Once my company was acquired by Hitachi and worked as a product manager there for about, you know, 15 years. The last five years they were deeply interested in building solutions which were machine learning, AI, big data analytics, space.

And that's how I got into AI, because the people who we were hiring, they were all data scientists and I needed to really learn that whole space very quickly.

And in: 2020

And in doing so we had a kind of, you could quote unquote call it corporate VC internally focused function as well. Because we needed to invest in only those ventures which have potential.

So we were reviewing those, we were funding them with the people as well as money.

logy fund and I joined him in: 2020

Andres Sandate: 00:07:30

Fascinating, you know, real deep backgrounds. I'm going to come back to you Al, maybe for, for one more comment on just the background and the team.

I know that Cordis is certainly more than just the two of you. You bring decades of experience, each in your own right.

And we could spend a lot of time on today's conversation, you know, just talking about things like setting up your first fund and being now a mid stage emerging manager in the, in the VC context. But there's so many other areas I want to dig into.

But I'll come back to you, Alan, maybe you can give another comment on the organization and the team before we jump into some of the themes and some of the areas of the organization strategy that I think our listeners will be really interested in learning more about today.

Al Tarar: 00:08:13

Thank you, Andres. So we are very serious people when it comes to investment management. So for us this is not a hobby, this is not a gig.

This is a professional fiduciary responsibility. So our current team consists of five people. All of them are very senior people like Rizwan and myself.

We have one woman partner and four men in our core team. Each one of us have 20 plus year experience on the average and graduate degrees.

Two of our partners have PhDs, one partner has two PhDs, one has one PhD and a CFA. So that's our core team. We do all things within our quarters. Among the five of us, we don't have younger junior staff yet. It's by design.

We wanted to make sure that we launched the firm with the right senior caliber knowledge and experience first and also affordability. As we get a little bigger we will add more people. On top of that we have about six to eight part time partners and advisors.

These are people that we reach out to for discrete question about a particular topic. And that could be capital markets, could be particular technology, could be healthcare, you name it. So we have that ecosystem.

And thirdly, we have a very large global network of contacts, former colleagues, friends, people we know personally and professionally. So we reach out to for broader kind of knowledge, information and experience sharing, so to speak.

Andres Sandate: 00:10:02

Yeah.

Which is so critical in not only the Bay Area and Silicon Valley, but just overall in terms of originating deal flow, doing due diligence, sourcing talent, references, teams, that human capital and that network effect is real.

But I want to talk about AI and, and the tech focus of the business itself and where you're at, what you see in the mainstream financial press and news is around what you said is generative AI. Can you just level set where we're at today with Cordis and where your focus is.

Al Tarar: 00:11:00

First, the types of AI. The way I see AI is embedded in my experience.

You know how I learned AI in the 80s and early 90s, I'm documenting a layman's or every person's kind of everyday language. Right. AI that we have been been learning and practicing for decades is pattern recognition.

This is how we as human being, as children learn things from different colors. How to recognize colors. Right. How to recognize hot from cold. That's called pattern. Right. Because we touched it.

And our parents would say don't touch it, it's hot. And we as children, we want to touch things, so we touch them and, and we learn that we should not touch heart. This is a pattern recognition.

So pattern recognition is very core to animal intelligence, particularly human intelligence. Right. Recognizing, experimenting. And then over time we know day and night.

We are 100% sure when it's day, when it's night, because we have seen it so many times. Second is rule based decision making.

Rule or equation or formula or law based, this is how we make a lot of decisions like in accounting or in mathematics. A plus B is C. So we learned that those are formula. And once the formula has been developed, has been proven, we just take it for granted.

We never question it. Similar to laws, they could be city, state, federal laws.

Once they are written, they have been written by elected officials, they become law and police and other people in courts apply them. But there is, that's a rule based kind of decision making.

So these two have been around forever, since human have existed, except it became a science 40, 50 years ago. People started saying how do we as human make decision? And how can we make it into computational kind Enough science.

So these are the kind of AIs that I learned and I practice since mid-80s, early 90s. Right. And they've been around longer than that. What we are hearing a lot about recently is artificial generative or artificial general intelligence.

It's basically taking those two methods of AI and adding layers of data and complications. Now they can make Very complex decisions.

Complex decision could be writing an entire report, writing an entire story, creating a brand new picture and so forth, so forth. So that's where the AGI is.

The next one which has not been fully developed yet but a lot of debate is happening that it will happen sooner than later is called artificial super intelligence. So basically there will be systems or technologies that will be so advanced they can actually make better decisions than human beings.

That's as scary. Right. But it also has the potential of solving some of the most difficult problems.

Andres Sandate: 00:14:51

When you bring it back to the wealth management context. I think a lot of investors are trying to understand how do you think about investing around AI?

What are some of the things that we can provide to our clients? Around less ethics, less politics, but more.

Al Tarar: 00:16:29

Well, I think Andres is in the academic arena. A lot of people are saying a lot of things. Sure. Honestly, some of them are so far fetched even for me and Rizwanapur will probably agree.

So to humanize and to put in a commercial context. Right. I'll tell you how Goldman Sachs chief investment officer looks at AI. He thinks, and I think you and I spoke about this a few days ago.

He looks at AI in a kind of a four layers or. Four. Yeah, four levels. One is hardware. This is with Nvidia. Like companies are building chips that are empowering computational technology.

A lot of your clients will probably be investing in Nvidia. People in general are investing heavily. So that's one. Second one is where a lot of chat GPT like companies are popping up.

They are developing very large language models. Basically they are studying everything that's available on the Internet.

Can you imagine any one of us reading everything that is available on Internet in order for us to answer some questions? That's what the second layer is. The third one, this is where we are focused is called AI enabled or AI based revenue generation.

How to do business, how to make money.

This is where we are and this is an area where AI has been very, very powerful as a tool to help doctors make better, more accurate diagnostics, patients get better outcomes.

Andres Sandate: 00:18:22

Yeah.

Al Tarar: 00:18:22

How to get better investment decisions made by advisors not to replace but adding assisting accelerating decision making of human being. The fourth level, then I'll stop is AI enabled efficiency process improvement. That's the fourth level.

Andres Sandate: 00:18:46

Just want to jump in here. When you, when you hear al break that down in terms of the four areas, the third and fourth areas, the revenue generating a revenue enabling.

When you hear that, can you give some more concrete examples.

Rizwan Muhammad: 00:20:06

So far before this whole chatgpt kind of came into the picture and you know, the work on on natural language processing kind of started. If you look at the technology, mostly it came from small companies, startups and you know, that's how you suddenly change the world.

But this time it's slightly different. You know, it was the Googles of the world who are investing heavily into AI way earlier than any small startup actually came into into the picture.

ittle bit. So al talked about: 1970

They there were really two reasons. One was in order to train any kind of neural network you need, huge compute power didn't exist.

And second thing was data and tagging of the data that was unavailable.

So if you wanted to build LLM back in 80s and 90s, you didn't have these two things you couldn't really do became possible after the Internet came and the world wide web actually became enabled and all the data started getting accumulated. And then the companies like Nvidia came because of a very different reason. They were kind of working on the graphics chip.

I mean this is how they started. But they were able to provide the compute power which was needed for the training purposes.

And where were all those things available in big companies, Google and Amazon and all those places.

So it was very natural for AI to begin over there, which is slightly different than any other technology you can look at IoT or any of other technologies we have seen, even Internet. I mean they were all started by small companies. So that's a fundamental difference this time.

That's the big guys were coming in, they had the infrastructure, they had the data and they were able to spend huge amount of money which was needed for it to kickstart. And that's how it happened this time.

So that's a fundamental difference in this technology now hype we have seen, you know, no matter which technology you talk about, there is always a hype cycle which you go through right in the beginning it's not kind of working. And then at some point people become overly excited about it.

So there is a, I would say more than a technology hype, there is a business hype always right? People think they can make millions and so people start investing in those technologies.

At some point you realize that, you know, wait a minute, there is limitation to everything. It's not exactly what we imagined it to be. And then you come down and the whole hype cycle actually goes on and on. So I think we are Kind of there.

The reason why so much money is suddenly getting invested in this space is because of what OpenAI did. They took the research work which was being done at Google, Attention is all you need.

The transformer concept, which was basically presented and they pumped in a lot of money and used it for natural language processing. And they were themselves surprised.

If you really listen to the CTO of the earlier cto, Eliar, he himself, when you listen to him, he says he was surprised the kind of results they were getting after the training. So that's why the hype got built up. It is amazing.

I mean, if you really look at you talk to ChatGPT, it gives you answers now which are quite exciting and interesting.

But at the same time, when you look at these LLMs which are being built for the health care purposes or for some specific purposes, people say for health care, for example, LLMs are not there yet. There's a lot of progress which has been made, but they are not there yet.

So slowly people are realizing where the technology is, how much of it is hype, how much of it is true.

At the end of the day, you know, building up an infrastructure, of course, you know, some of the companies are going to benefit from that, which is Nvidia's of the world right now. But at some point that cost has to be recovered.

And that's where my worry is, that, you know, if you really look at amount of money being spent on the infrastructure and if the, the top vertical solutions which Al talked about earlier, they don't come in and they don't start, you know, generating revenue, who is going to pay for it? I mean, right now Google is paying for it, Meta is paying for it. But at some point it has to make a unit economic sense.

Al Tarar: 00:25:13

Closing the loop on a remark Rizwan made earlier that pretty much all major innovations in the US have been created by smaller companies. So these two first layers, we talk about Nvidia, which is hardware, and then large LLM models.

That's where Microsoft, Oracle and the likes are putting billions of dollars.

But the vertical solutions that we talked about, AI based revenue generation and AI based efficiency, those will be done and they're being done by smaller companies and those are the companies that we invest in. So they are not trying to change the world.

They're saying, I want to change a particular part of healthcare, I want to transform chronic disease management, I want to transform how people learn in schools and colleges, how to make it better, things like that. So they're fixing a narrow but very specialized problem.

Andres Sandate: 00:26:14

I think the areas where, where you're, you're focused, education, health care, we've talked about security, Security, right. Cybersecurity, logistics is a big area. And then I know that Fintech is another big area that is, is going to be included.

So across those four, that covers a very large segment of our, of our GDP of.

And what you're saying is that there is going to be vertical category winners if you will talk a little bit about why they continue to need capital and where your stage of capital more important comes into play.

Al Tarar: 00:27:02

So we tend to invest in companies that have been around for on the average five plus years.

They have developed product, service, they have been able to sell successfully to commercially paying clients and they have meaningful revenue and their revenue could be a million plus, could be 5 plus million on the average 3 to 5 million dollars annualized revenue. And that's what we call early growth. So they have been around now they are about to grow. So they need money to grow their business.

So we, that's why we call venture growth investment. In this investment cycle the risk are much less, but upside potential is still quite significant.

Actually Cambridge Associate, which is a $300 billion investment firm, you know them, they call this strategy as the best risk adjusted return strategy. So this has the upside of a traditional early stage venture capital, but it has downside risk protection of a more mature businesses like buyout.

So that's where we invest now. What some of these companies are doing now would be unimaginable few years ago thanks to the innovation in AI.

So one particular company that we talked about is AI based chronic Disease management.

So chronic disease by definition is a disease that cannot be cured and hence this needs to be managed, you know, chronically over the, the life of a patient. And examples are diabetes, you know, dementia, Alzheimer's, etc etc.

So for patients who are going through these chronic diseases, let's use dementia, early dementia as an example. These are Medicare patients, mostly 65 plus year old.

They take on the average 10 plus prescription medicines and they have been on prescriptive medicine for decades, years to decades. The data that has been accumulated for that single patient is millions and millions of lines or brackets or data points.

An AI enabled software can go through that in the matter of minutes versus a doctor going through this.

Can you imagine a neurologist reviewing 30 year or 20 year or even 10 year worth of data of a patient and trying to understand what this patient might be going through in terms of dementia? AI can help the doctor do that.

This company that we are investing in is using DNA, blood work results, prescription medicine history and personalized kind of lifestyle questionnaire. They take all that and AI model generates a report for the doctor. But this report is different than what a large learning models will generate.

It's not generated by model, it's generated by a more complex AI system. Everything in that report is footnoted with a fact and science. So everything in that report is defensible or defendable by the doctor.

That's the difference. That report is being generated in three minutes. If a doctor had to do the same work, it will take him or her weeks. Sure, you go through the data.

So that's where the significant value is created by this kind of technology company. This is just one example.

Andres Sandate: 00:31:37

I want to ask you when, when you hear that example from Al, which is you know, one from the portfolio or you look at a prospective investment opportunity as, as Al said, these are businesses that have, they've got recurring revenue product market fit, they're now looking to grow.

When you look at these businesses across the verticals that you're focused in, how do you go about filtering beyond just looking at the revenue numbers?

Rizwan Muhammad: 00:32:13

Yeah, so it's our investment framework as such, the way we analyze these companies that foremost we look at if there is a real business problem they are trying to solve. I mean that still stays, right?

Andres Sandate: 00:32:27

Yeah.

Rizwan Muhammad: 00:32:29

After that we start looking at the technology, the deep tech, what they are using, how they are using it. So we know there are four or five areas where AI right now is actually hot or is much more applicable and everybody knows about them.

There is the fintech space you talked about. There is healthcare, there's education. Right. So at least those three are on the top. And then there's industrial.

So we kind of analyze our companies which we are looking at from that lens. And so you talked about, I mean I was giving an example of a health care company.

So this company essentially still uses in its core the rule based engine. It's a decision support system at the end of the day. So in each and every area and field, AI's application is going to be slightly different.

What problem you can solve in each area is different. So in healthcare we know that, yeah. At some point the CN sport systems could be leveraging LLM but LLMs are not ready yet. Right.

So we are very careful from that perspective because we don't want to be investing too early in a company which is actually using that as a technology. But as long as they're using AI, which is rule based engine, decision support system, we are good.

Now there are spaces, there are areas where they can leverage the modern generative AI as well, for example, as a part of analyzing the research papers and extracting information from their building rules out of it, which is a natural language processing problem. So are they leveraging that? So that's what we look at, right? Yeah.

So similarly, classification part, when you look at identifying or diagnosing cancer, the binary classification, whether you have a cancer, not have a cancer, those neural networks are available. They are pretty good. They are better than humans.

This is how we basically, on case by case basis, we analyze that where this technology is being leveraged, where it is being used, whether it can actually be used over there or not. Are we too early or too late from that perspective? So it's art plus science.

There is some methodology which we use to filter out things, but at the end of the day, it's really our background. 25, 30 years in the industry, you have seen things, you've seen hype. I mentioned earlier.

So if you, you understand the technology, you can sit down and see that whatever problem, quote, unquote, business problem, is that valid. If they are trying to solve that, how they are solving it, Is it a solid technology underneath, or is it, or is it.

Al Tarar: 00:35:14

And we can see height so we, we don't touch a blingy thing. We call it, you know, shiny things. Everybody's saying this is going to be the hardest thing to like.

If somebody comes to you as an investment advisor saying, as using an example, saying, andres, I have the best security that, you know, you and your clients should invest in, would you invest in? Probably not, right? Yeah. It's like if this is too hot, it is too hot.

So we don't invest in too hot an area, too hard a company because of obvious reasons. We want to see this is mature enough, this has been established enough.

There's enough data to substantiate that this is a good place, good area, good sector for us to be in.

Andres Sandate: 00:36:16

What we as advisors are getting all the time bombarded is again to your point, a lot of investment ideas. I'm sort of skeptical because how much of it was AI before and how much it is now AI because of marketing.

And that's what the marketing team suggests should be in there because it's a buzzword, right. And, and, and, and it's going to catch people's eye and attention, right?

Al Tarar: 00:36:37

I absolutely.

So I think one thing for your audience, another piece that I recently read in Financial Times, one of their reporters, she came up with the idea to humanize this artificial intelligence. So she said, why do we call it artificial intelligence?

Why not call it, you know, augmented intelligence or accelerated intelligence or assistive intelligence, basically assisting human being to make better decisions.

And so it's just a word, but there are serious concerns around the safety and ethical aspects of AI, which is being looked at by many, many people around the globe, including us. I want to make sure we convey to your audience that there are a lot of safety measures built into all practical application that we are investing in.

We do not invest in any area that has not been well established or proven beyond reasonable doubt. You know, and we have convinced ourselves that this is a commercial idea. This is not a academic or scientific idea.

Andres Sandate: 00:38:19

It seems that it's more focused at the commercial level. Right. Healthcare, education, there's stuff that's for built for industry, built for enterprise.

And that sounds like where, you know, Cordis is really focused.

Al Tarar: 00:38:59

Yeah. We are absolutely not a B2C.

Andres Sandate: 00:39:02

Right.

Al Tarar: 00:39:02

So by design, yeah, there's nothing wrong with B2C, but we don't do B2C. So we are B2B because we believe that is a more mature, established, safer, you know, and we understand that area better.

Andres Sandate: 00:42:46

When you think about the Cortis fit, how do you begin that qualitative quantitative march to get down to ideas, to the things that you want to spend more time on?

Al Tarar: 00:42:58

I will let Rizwan answer second part, but I will add a front part first. So think about this, Andres. We are actually benefiting from many other funnels first.

So before the companies comes to our funnel, that company has gone through 2, 3, 4, 5 funnels already. Do you follow me?

Andres Sandate: 00:43:22

Yeah. In terms of the rounds of capital they've already raised.

Al Tarar: 00:43:26

So it's just like we are benefiting from the company's difficult life or the time horizon they have gone through. By the time they come to us, they have meaningful commercial revenue.

Technology has been established, solution has been established, yet we go through a very deep due diligence, which is one of the common.

Rizwan Muhammad: 00:43:55

Right. So he's. Al is right. I mean, still, although, you know, there's.

There's that filtering going on that we are, you know, investing mostly in the CDC Series B, but there is other filtering going on. The people who are sending us the deals, they are very careful in terms of what they send us. Right. But then still, we get so many deals.

You know, it's maybe 10 a day or more than that. I mean, that too, after filtering. I mean, we are. So there is an email filtering going on between L and me.

If we are getting it directly or if people are submitting. But beyond that, the mechanism which we have established actually is a multi stage process.

So we do initial filtering which is based on that criteria which we have set up for our fund, essentially that we are going to be investing only mid stage companies which have certain amount of revenue and they are in these three or four spaces. So that filters out a lot of stuff.

Andres Sandate: 00:44:51

Sure.

Rizwan Muhammad: 00:44:53

Then we also are aware that we are a small fund and so we cannot invest in those areas where huge amount of capital is needed. I mean robotics, for example, we love that space. We know artificial intelligence is going to have a huge impact on robotics. But we are small fund.

This fund, initial first fund is small. Eventually at some stage we'll open it up because we would, we love that area as well. So it filters out some of stuff like that.

And then based on my experience at Hitachi, because we were building some sort of a very easy to use process and mechanism, we have actually leveraged some of those concepts and built our own scorecarding mechanism, which is based on nine or 10 metrics, which are, you know, the financial metrics as well as other metrics. And that's what we use that you were talking about the comparison, Right.

So we generate charts out of that which actually help us compare companies against those which we already invested in, where they fall in the category. And that actually provides a great discussion point internally. It's just a tool. It's not a, you know, the final decision maker.

It's a decision sports system, if you want to think from that point of view. So that actually helps us do the comparison. This is only those companies which have come through the filter. Right.

Then after that, once we identify a couple of those which are actually really at the right spot in terms of the charting, which we do, then we do, then our due diligence actually starts the deep due diligence, right. Which is a long process. Looking at their data rooms and so on.

This is where the earlier conversation which we had that depending upon which area that company is, we deeply look into the technology where they are using or leveraging AI, how they are leveraging it, whether it makes, I mean, we look.

Al Tarar: 00:46:47

At patents, we look at how they use AI, what kind of AI they are using, how sustainable that is. So it's, think about it, capital efficient companies. So capital efficiency is very important.

And we are trying to create a concentrated portfolio of 10 to 12 companies, not 100 companies. And out of these 10, 12 companies, we are actively engaged in helping these companies improve their speed.

Of growth, their strategy, their market access.

Andres Sandate: 00:47:22

Yeah. Because you probably want to land them, as you say, and then expand. Right.

You want to get going with these companies, but then as they grow, you want to participate and be a part of that growth that's going to benefit.

Rizwan Muhammad: 00:47:35

That's why the management team is part of that nine metrics which we look at, Right?

Andres Sandate: 00:47:41

Yeah.

Rizwan Muhammad: 00:47:41

Because we are going to be working very closely with the management team. We want to see if they are mature enough to be able to go through difficult periods. Every company is going to see those difficult periods.

Can they pivot? If there is a need to pivot based on how the market is changing. So all those things actually are bit of art as well. When you talk to people.

Andres Sandate: 00:48:22

You talk to a venture manager, whether they're early, early stage or they're mid stage. Risk management and margin of safety don't often come up. For us as a wealth advisor and a fiduciary, we're looking for diamond in the rough.

Portfolio managers and emerging managers for sure, we love because we want to find those teams that are highly incentivized and motivated early in their life cycle.

We like to find the teams when they're early, fund one, fund two as the emerging manager because there's a lot of skin in the game and they're generally very, very focused on getting to the next fund and very focused on returning capital. So maybe leave you a chance to comment on margin of safety risk management in the context of early stage technology investing.

Al Tarar: 00:49:12

So, Andres, I think for early stage venture capital, your statement is absolutely correct.

So a lot of early stage venture capital firms would say we are being hired to take risk, we're taking risk, we are investing in 100 companies or, you know, 20 companies and one or two of them would be successful and they will make it worthwhile. Our strategy is the opposite of that. So our entire thesis and strategy is that pretty much all of our companies should be successful.

Not one, not two, not majority, all of them.

Now, the difference will be some will be more successful than we expected, some will be less successful, some will be successful sooner, some will be successful later. But the idea of completely wiping out the capital that we invested in a company that is not part of our, what you call a calculation.

So we don't say we are putting in 10 companies, we expect, you know, one company to black out. Now, statistically speaking, for this strategy, according to Cambridge Associate paper, the capital loss ratio is 13.5%.

Rizwan Muhammad: 00:50:30

Yep.

Al Tarar: 00:50:30

What that means is it's not the idea of 13.5% companies will fail. That's not the idea.

The idea is after you complete all the portfolio investments, you might lose up to 13.5% of the invested capital, but your remaining 80 plus percent could generate 400% return. So that's why Cambridge Associate concluded in their study that this is the best risk adjusted return strategy. And that's what we like it. Right.

So very risk management focused strategy. In everything that we do, we try to avoid the downside risk.

So we have very severe protections in liquidity, liquidation, preference, priority or seniority. Then we have the participating preferred shares. We never ever invest in common shares.

So we are typically the first to receive the capital before anybody else receives if we are series C, Series D. Right. So there are many different steps that we take to what we call managing the downside risk.

Andres Sandate: 00:51:50

Yeah, I mean, structure matters.

I mean, so much of investing at the diligence that we've conducted, you often can find a portfolio management team that are great at sourcing, but they might not have been cycle tested or they might not have been through, you know, setting up a fund before, or they may not have the breadth and depth of team. Right. To really back up the claim that they have the network.

All these variables come into play when you're doing the diligence on the manager and the management. When, when you think about the, the background you have, where you're at today at Cordis, what gets you most excited?

Rizwan Muhammad: 00:54:12

r and then Covid was there in: 2019

What we have seen is, for us at least, I mean, I'm sure market is still going through that difficult period in terms of fundraising, but we have seen tremendous progress in the past six months. And that's what actually makes me very excited because want to get over this hump of raising the fund and focusing on the companies.

's what I'm very excited that: 2025

We are done with the capital raising and we are actually focusing on companies and helping them out. So that would be very exciting stuff.

And from the technology point of view, we know AI is becoming more and more prevalent everywhere in each and every field. How the market actually changes from that point of view. You talked about Big caps versus mid caps.

Mid caps definitely are making progress, which is a very, very positive, healthy sign.

I mean, I'm looking my portfolio in Fidelity and if I look at some of those mutual funds, last year the largest gain was in the mid cap mutual fund, which I have 60%. That actually tells you, right.

That tide has already shifted in favor of, which is actually very positive from our perspective because you know, these companies are going to be as, you know, the mid cap first before they become big caps. Sure. Opportunity is definitely there. It's, it's just a, you know, matter of executing now.

Andres Sandate: 00:56:16

Fantastic. All right, Al, you get to bring us home here with the final comment.

Al Tarar: 00:56:21

So final comment is for your audience. AI is here to stay. There are some scary potential prospects that many of us sometimes think about.

I would advise your audience to not worry too much about them now because a lot of people are looking at those negative or severe possible impact. They are possible, they are not necessarily probable in the near future.

Governments, agencies, people, professionals like us focus more on the positive, constructive benefits of AI to many human endeavors at work in business, in law and security and healthcare, education. That's what keeps us optimistic, that keeps us excited, motivated to push forward.

Andres Sandate: 00:57:35

Really appreciate the chance to talk to you. Al Tarar and Rizwan Muhammad, founders of Cordis Capital Partners.

They're focused on the growth stage of venture capital with a technology and really an AI focus. Al, what's the way that folks can learn more about you guys?

Al Tarar: 00:57:50

Our website is quartuscap.com q u a R T U S c a p1word.com so if you go to quartascap.com, you go to our contact page, you can send us email, you can subscribe to our materials. So we welcome any and all people to reach out to us.

Andres Sandate: 00:58:13

Well, we'll leave it there for today. Gentlemen, thank you for for joining me today on on the podcast. It was a fascinating conversation on AI and what you're building.

I wish you guys good luck in finishing up the fundraise for your first fund and I look forward to having you back on the show perhaps later in the year to get an update and hear all the developments in the portfolio. But we'll leave it there for now.