Rajan: Welcome to Pivotal Clarity, an AI podcast. Our guest today is Arshad Hisham. Arshad is building the future of AI and robotics as a founder and CEO of Ingen Dynamics with a vision of building intuitive and practical robotics and AI solution. Ashid is not just another tech CEO. He's an inventor, engineer, and educator all rolled into one. And his company recently raised overdose, uh, and $50 million. What intrigues me most about Ashad is how he's approaching AI development holistically, not just focused on the technology, but on cultivating the next generation of AI professionals through initiatives called futurenots, uh, program. We'll dig more into that as we speak. Um, I'm very excited for this, uh, episode, uh, to learn from Arsher journey, and, uh, here his thoughts on where AI is headed. Let's get started. Arshad, welcome to the show.

‍

Arshad Hisham: Thank you, Rajan. You know, pleasure to be on the show and happy to contribute.

‍

Rajan: Yeah, let's start at the beginning. What made you decide to found Ingen dynamic? Was there a specific problem you encountered, or was there an observation that sparked the idea? I'm curious to know about your thought process.

‍

Arshad Hisham: Sure. Uh, I think, uh, Enjin Dynamics is my. Just to give you a bit of a background, it's my fifth company, so I've started five companies since my, uh, business school, and my companies have grown in complexity. And I started engine Dynamics after I exited my last fourth company in 2015. And 2015, I think there were all the right ingredients for revolution in AI, automation robotics. I think it was a great year. The cost components, data, uh, cloud was decreasing exponentially. So I kind of figured this is a great time to do something in AI, automation robotics. And, you know, this is where the future would be. And, uh, I think the last eight years have proven me right if. If I would like to kind of. At that point, I think, you know, we're kind of heading in the right direction. And there, uh, are a lot of exciting opportunities happening in AI, automation robotics. And to go back further along, I am a big science fiction fan as a lot of tech entrepreneurs. So, uh, you know, uh, can you just give me a minute? Rajan, this is a delivery I talked about. Yes, sorry about that. Just pick it up.

‍

Rajan: Go ahead.

‍

Arshad Hisham: Yeah, so, I am a big science fiction fan, as with most tech entrepreneurs. And, um, early on, when I was eleven years old, old, I read a lot of Asimov's foundation series. You know, it kind of fascinated me what a future would look like with humans, robots, AI, all in the mix. It was very fascinating. So I was driven very early on from a young age to bring a lot of these ideas to, uh, reality. You know, that kind of drove me throughout my career, and I was just, uh, you involved in a lot of projects where I got a lot of experiences. So this is kind of really the starting point. So that's kind of, uh, in a nutshell how I ended up founding ingen dynamics.

‍

Rajan: Uh, let's talk a little bit about origami, the platform that you're building. It seems to be the core of what you're building at, uh, ingen Dynamics. Uh, curious about how building a platform works.

‍

Arshad Hisham: Sure. So the idea behind that, as with all companies, uh, in hindsight, it's easy to connect the dots. But when we started out, uh, we were creating a robot product which had a home robot platform. We were never had a platform play in the mind, but as we build the different components, we build out the product. We got some feedback from a couple of accelerators in the US, including plug and play, where we got feedback from a lot of companies. Uh, we started creating various products specifically for healthcare. And then we realized that, ah, the set of components that we had is kind of reusable in AI, automation, robotics. And that's why it's called the origami platform. So it's like a trunk, you open the trunk box and you have different components, and you can rearrange these components and create different products just like you create the paper origami. That's why it's called the origami platform, by the way, just like the paper origami, you can recreate. So this right now is at the core of the company. It contains all the components that you build. It will be a future intellectual property for the company as we grow. And our vision for the company is that for the first two years, uh, we have a lot of dependent hardware reshaping for the next two years, or phase two, which could be more than two years, is that we become hardware uh, agnostic, more software specific, software bundle specific, and phase three, we become software and hardware agnostic. We become a pure play platform company. So this is the vision that I, uh, have for the company. And then third party providers could pretty much, uh, use our platform to create their own products. And, you know, so this is, this is a vision, but it's very easy. I mean, the track with a lot of platform companies is that it's very easy to say that you're building a platform and platforms takes a lot of time to get traction, you know, to get the inflection point it's a uh, very hard thing to build. Now all this runaway success that you hear about has been that, what I call them, the ten year overnight stories, right? This is not easy to platform. So what we are building here is actually we are uh, creating the use case for the platform upfront, but we are showing that the platform can exist and then we open up. This is a vision that we have at NJ. So that's why you see this unique strategy of the three phase strategy rolling off platform.

‍

Rajan: So you said this is uh, like an evolving journey. But I, let's say somebody is uh, getting started right now. And when you're building and uh, selling AI like uh, what's the best way to get started? Maybe you start it as a service. Do you build out a product and then think about platform or now what, you know, is it better to sort of start uh, from building a platform, uh, from the right, uh, at the start?

‍

Arshad Hisham: I think I would still uh, advise for the near term use cases. I think the platform is definitely a longer term play and this is kind of where a lot of the opportunities lie. And even, I mean the unique thing with AI, as you very well know, is at the speed at which things are happening and developing, right? I mean it's so rapid and there is always this, you know, I tell this a lot to my other founder friends that I'm always waiting for the next OpenAI release to see how many startup business model kills, right? I mean, you know, this is so crazy. So there is always a risk that your whole business model is become obsolete by the next open AI or large model. So I think a uh, smarter strategy for entrepreneurs. I think there are two routes now for entrepreneurs. Either you go the very long run, which would need a lot of capitalization, which is a big boys club, or you build a product layered around that which has some unique services. And this is not, they're not just using LLM under the hood. I think those are not sustainable. A lot of the startups use an LLM under the hood which runs. So you need to create a secondary ecosystem which would be kind of building around this, the primary ecosystem. And that's where I think there's a lot of fantastic opportunities for enterprises and business. And if you can scale that, I think uh, it's a great way to build. So that's my advice. My advice is that you should concentrate on building out this use cases and try to scale that up.

‍

Rajan: You talked about uh, platform games being like big boys Club, a lot of fundraising, you recently increased your investments to about $150 million. That's a lot of money. Uh, how do you think about allocation of capital? How do you actually split this fund? Uh, how should one think about it?

‍

Arshad Hisham: Sure. I think, I think the um, see the capital allocation is more towards uh, so we have a post IPO strategy, right. The game doesn't end, the game only begins at an IPO, right, for us or the liquidation event. And then the real game starts. You know, you have quarterly pressure, you need to meet the numbers, you need to meet the growth. So a lot of this reserve, as we kind of grow post IPO and figure out how do we kind of get to the right points. How do we kind of get to the right, uh, uh, right inflection points. So a lot of it is for growth, a lot of it for R and D pretty much, and distribution. So distribution, I think distribution is a new competitive advantage. As you probably know, distribution is key. How do you distribute, uh, you know, your products and what synergies you can bring. So I think it's a long game and given the, given the new funding scenarios, I think I was reading somewhere for the new round that OpenAI did, the minimum size for VC's to get in on that round is 250 million. That's a minimum size. Now just want to let that sink in for a minute. That was a total fund allocation size a decade back for one single fund across multiple portfolio. Now that's a minimum size for the opening as latest round. So I think the capital game is changing and that's where I think you need to be very conservative with how you look at your budget, how you kind of figure out uh, where to go in on the next steps. So this is kind of, kind uh, of a key thing, right? So we are playing along. So we have a clear post API strategy where we allocate different budget sizes for different use cases. And you know, that's, that's a short and long answer.

‍

Rajan: Uh, you are building robots to fill jobs. That's a big order, right? So what kind of jobs are we talking about? Let's say I run a hospital or a hotel. Why would I want a robot instead of human workers? What's the real advantage there?

‍

Arshad Hisham: Sure, great question. I think um, there are certain targeted healthcare use cases where you know, there is an acute labor shortage. Classic example is elderly care. Okay? I mean there is a, ah, massive and acute shortage in Europe, in North America, in Japan. Uh, if you know these countries very specifically, they've been doing a lot of R and D into elderly care. Lot of research into elderly care. So our vision for the robots is that we have simpler form factors in the near term, which are targeting the low hanging fruits over five to ten years. And then we have complex form factors coming in, like the humanoid and Rovers later on in the longer term, right? And all this is powered by the underlying origami air platform. So the platform keeps on building up competency and we power it. So this is. We call this the phased approach. Right? The phased approaches. We call it because we have simpler form factors targeting near term use cases, including healthcare, telemedicine, elderly care, uh, restaurants, in terms of hospitality, there's a lot of shortage there. And then we move on to the complex form factor. And then it gets interesting, because if you talk about humanoids, and I say this to a lot of, uh, people, that using humanoids is actually the ultimate hack. Why? Because we have created the world, the whole world for humanoids. Think about that for a minute. The escalators, the cars, the buildings is designed for humans, humanoids. So we have invested trillions of dollars creating an infrastructure. So the whole idea of creating a humanoid is not. See, people get confused. They think that we are creating a humanoid because it looks like a human. The answer is no. The minutes are created because it's a quickest hack to access or tap into a trillion dollar investment we made in assets over the last hundred years. So we create a humanoid form factor. We can tap knobs, cars, escalators, everywhere, right? And humanoid is a massive disruption model because once that comes in, you know, we could be talking about covering up at least 30% of the GDP could be powered by human rights. It can be anywhere, right? Manual labor, specifically manner of labor. And, you know, once that goes in. So there is a lot of destruction coming our way. Right? So this is, in a nutshell, the roadmap. So hope you get an idea, like how we are. We are rolling, uh, out products, but definitely our vision to create a better future and a safer and a better future for humans, right? So we want to improve the quality of life. We want to free up people to do the things they love the most, right, and not get burdened with boring, repetitive, and dangerous jobs. So that's where I think these robots will play an important part in freeing up, uh, productive and leisure time. So, humans down the line, uh, you.

‍

Rajan: Spent a lot of time, uh, working in AI. Can you share an example or a time when you saw someone using AI, uh, in a way that you didn't expect?

‍

Arshad Hisham: Sure. Just to give you a bit of a background I published my first paper in AI in 90, 819 98.

‍

Rajan: Long time ago.

‍

Arshad Hisham: Yeah, yeah. And it was called, it was not even called an AI. I remember it was called the component of back propagation network. And we used it for my robot soccer project at NIT. I did my undergrad. So we did a project on soccer robots and we used it to predict uh, the direction of the robots. And you know, we had to train the model, use the back propagation and it was fascinating. Right. So I think very early days it was not cool. It was, it was around the same time that you remember when I uh, when the first digit reading ML algorithm came in was around 98. This is around the time. So that is a pivotal point where things really took off. But down the line, I've seen several interesting use cases, uh, for AI. I think one was uh, predicting, predicting the office commute was kind of an interesting use case.

‍

Rajan: The traffic problem.

‍

Arshad Hisham: Traffic problem, yeah. But I think it was on a very personal note, somebody used it to predict, you know, uh, a local problem. And it was very, very innovative. And if you see most of the inventions have been created by very smart but very lazy programmers and inventors. They are the real smart cookies, right? They invent some tech because they're very lazy. They don't want to kind of waste time. So they invent something that is, that's good. So this is kind of where I think, you know, this has happened. Sorry, Rajan, I think I have one more delivery. It's a new hotel. Yeah, I'm just going to pick it up and promise this luck. Yeah, go ahead. Yeah, so I think we were talking about. Yeah, some of the best use cases of AI I've seen are people who are like very smart, inventive, lazy inventors who kind of created uh, a solution to their day to day problems. And that's where I think you see a lot of innovation. And that's, and that's true for the past, the pre AI era. I mean if you remember the 20 1090s, it was true then. If you remember the windows, uh, developers and hackers and the Linux developers, we found the best innovation from this group of people. So I've seen the best use case of AI from similar set of people.

‍

Rajan: Let's say you're having dinner with three people from history to talk about AI and creativity. Uh, who are those folks? Uh, what would you ask each one of them?

‍

Arshad Hisham: It's a tough one. Uh, so I think uh, one would be obviously Alan Turing for sure, I think. And uh, the question would be like whether he can predict the direction that AI is making right now and whether it's Turing test will still be applicable because I think we are far reaching a point where the Turing test is no longer sufficient.

‍

Rajan: Very interesting. This is the second, uh, guess that I'm hearing he say that looks. Looks like all of us want to go back and talk to Turing about the Turing test.

‍

Arshad Hisham: But I think. I think that's. That's a, uh, uh. That's a, uh. That's the first person I want to. I want to talk to. Right. I think in terms of AI and. And all the other people I want to speak to are currently such a new field. Are currently right now, you know, involved in all this process. Right. So there is kind of such a developing, uh, developing field. You know, I'm obviously fascinated by a lot of work by, uh, Mustafa, uh, Suleiman, I think, you know, from. From deep mind and, you know, and, um. Yeah, not just a tech side. I mean, he has written a book which is an important read for anybody, you know, which explores what are we dealing with, you know, what is the next decade is going to unfold as all this AI unfolds in the world, you know, about society. And so from that perspective, I think, you know, this is something that, uh, you know, so do I need to pick all people from the past or do. Can I pick. You can pick any kind of currently living and working as well?

‍

Rajan: Yes, yes, yes, you can do that.

‍

Arshad Hisham: So I think my question would be to predict the next ten years. And, you know, because he predicted in 2010 when they started DeepMind, the next ten years, which was kind of accurate and kind of, uh, kind of doing, uh, what to do. And my third person to speak to would be like, right, so Samaultman, uh, classic question is that he was seeing in, uh, speaking in one of the recent events that the mood slow is going to operate at not hundred x, 100,000 x because of the new changes in AI. So we're going to see a, uh, massive uptick in. In the Mosler in terms of processors and knowledge this for the next decade. So how would that. And this is so wild that I don't think we as, uh, a species can comprehend the scale of this transformation. Right. We are not wired to understand the scale of transformation. So how would the next decade look like with 100,000 x more slow in play for genetics, for AI and self learning and AGI, you know, so we are fast approaching. We can already see the seeds there with the latest version, the strawberry coming out. There are, ah, seeds of AGI already planted. It's already moving to that direction. So now it looks, there are a lot of bugs. Right now if you test, things are not working. But this is very similar to the early video generators two years back. It is very buggy, but now it's super smooth. I'll wait and see what, uh, the next two years fold. So my question, third person, would be to some outman to predict the next ten years with 100,000 x more slow, what's going to be.

‍

Rajan: I've seen you talk about this thing called, uh, zero corporate bloat and being ethical. Uh, what does it mean in practice and uh, why is it important?

‍

Arshad Hisham: Sure, I think, uh, yeah, I think a lot of time as uh, companies grow bigger, you tend to accumulate bloat, which is buffer. And you know, there is inefficiency. Grow, not intentionally. Uh, it tends to happen because the incentives of different actors in a company changes as a company grows. Right. Some enter into kingdom build. I call it the Empire Building. The executives start building empires, you know, power. They tried build up silos and the goal is to maximize, which is not wrong, because the problem is many of the companies are incentivizing that indirectly. Right. So this is what, what tends to happen. So over here, I think for me, my vision is clear. I think I need everybody to be aligned with, uh, zero corporate blown. That means it's as efficient as possible, you know, as a agile startup. And there were a lot of recent talks even, you know, I, I wrote this out probably like four or five years back, but very recently, there's a lot of talk about founder mode. I'm sure you've heard about the founder mode talk that's going on. I think it's very similar. So this, yeah, so this is a founder mode I'm talking about. Right. So I think I, and I had practiced this. I think it's very true. Absolutely true. Because, you know, I had to kind of change gears. I do kind of go down tires and kind of explain what's going on. And that has led to efficiency. So one way of cutting out corporate bloat and cutting out politics is to kind of operate at different levels, not just have a higher. And that's found a mode. And this is very true, because if you give your, if you give too much autonomy everywhere, I'm not saying it's, it's not micromanaging, but the problem is there is a tendency to accumulate bloat along the whole chain of command. And that's, uh, that is like a law of physics, you know, that is where the natural direction is. You can't prevent. The only way to prevent it is activate founder mode every once in a quarter, go down and skill it up. Then you can reduce the blood, not remove it. You can make it like 10% rather than 60% as thunder process. And that's what I'm calling. I call that the founder board. The company needs to be running.

‍

Rajan: Um, let me do a quick round of short questions. You can just give me the first answer that comes to your mind. Um, a book that shaped your approach to AI. I know you already talked about one book, but is that the same book?

‍

Arshad Hisham: Yeah, I would definitely say it's a foundation series by. Yeah.

‍

Rajan: Uh, and it's the biggest misconception about it.

‍

Arshad Hisham: Yeah, sorry, go ahead. No, I just talked about Asimov. Yeah. It's a very unusual selection because it is not a textbook or it is more of a, uh, seed generating book. Right. So what AI could do? What? Yeah, so that way. Yes, that was my biggest book.

‍

Rajan: One big misconception about AI that irritates you.

‍

Arshad Hisham: I think this classic Hollywood stereotype of AI turning evil and, you know, destroying humans, I think it is wrongly phrased. I think there is a big existential risk for AI to destroy humanity, but not in the way that movies depicted. You know, this is, would be like the paperclip maximizer problem, where you design an AI to maximize paperclip, and it turns the whole world into a paperclip and accidentally destroys the human race. So this is a risk not, you know, the AI becoming jealous of humans and doesn't make any sense. You know, the most depiction of hollywood AI and robots doesn't make any sense. So that's what irritates me a lot.

‍

Rajan: What's your favorite productivity hack?

‍

Arshad Hisham: I call it, uh, I call it the 440, uh, kind of a thing, because I try to do at least four things in a day. I have a big schedule agenda, so four things I can fully complete, and I have a list, and then I keep on moving, uh, and have 40 printing items in the next slot. So this way I'm very organized. Every Sunday I keep my 440, and every day I make sure that I finish four, but more than four I don't target. And it's surprising because in a week, you can get at least 24 items done. If you try to do everything at the same time, you'll get only, like, five to six items done. So I think having this, I, uh, call the 440. So 40 items, um, pipeline four items to complete nitty. So this is the best thing that, and it doesn't matter. You can use any tool, excel or whatever, but follow this process. I think it will work wonders for you.

‍

Rajan: One word to describe the future of AI.

‍

Arshad Hisham: I think is, uh, fantastical would be the one word. You know, you can't even imagine it. It's going to be fantastic.

‍

Rajan: Tell me a little bit about the futurenauts program that you're doing. Why are you running it? Uh, most companies that I know, they just hire talent. Uh, you seem to be creating it. Uh, do you feel that it is worth the effort that you're putting into it?

‍

Arshad Hisham: Sure. I think futurenauts, again, was originally designed for training our engineers and, uh, in our team. So that was original intention. But then we had a couple of universities and engineering colleges that reached out saying that they want to launch a, uh, BTEC program with the futurenotes initiative. And then we started. So the reach out came from the industry and from the universities, and then we converted to an engineering degree. We had schools, uh, we had a partner who was doing school education, and we kind of interface with that. So it kind of started out, and if you see future notes now, currently is at the intersection of what we create internally and this external partners. So it's kind of sweet, right in the middle, and it's a great way to build talent. And there is a massive reskilling going on, because every job, this is a good McKinsey study, which said that at least 95% of the job would, uh, have more than 70% of their task changed to an AI specific subtask. So there is massive reskilling going on, and it's not. Jobs are being gonna be taken away. I think the jobs are gonna be redefined, so people have to be retrained. And this will be the biggest problem of the next decade. And what a, you know, what is a great way to address that is have this program and roll this out and, you know, make. Make sure that we have the next set of innovators coming out.

‍

Rajan: As a robotics company, um, how do you make money? What's your fundamental business model?

‍

Arshad Hisham: We have different models. We lease our products and we charge for services, so we have a monthly recurring model, and we can sell the data and, uh, software on top on a monthly basis. So this is kind of a key structure that we follow. But the harder thing as a robotics company is that, uh, it's a very capital intensive business. So there is no, um. It is not. You are used to the enterprise world. The enterprise SaaS world is beautiful. It generates cash and you can scale and you can predict and you can go back and iterate in a week. Robotics, it takes six months to iterate. And, you know, to iterate, you need to raise further cash, you need to pivot. So I think there are harder problems in getting to PMN so that, you know, allows for some, some different business models and different cash flow and different semblance of creating cash flow, which is what you can see on website. We are not a single revenue, uh, stream company, multiple cash flow sources going.

‍

Rajan: Just a few last questions, uh, Ashrat, um, uh, what questions do you think that people are not thinking about AI and robotics and they should?

‍

Arshad Hisham: I think one important question is, I think the biggest problem we're going to face for society, for government, for institution, is that, uh, the world is not ready for the massive upheaval that's about to happen in the next five years. And the government said, nobody's prepared for it. And I'm not, I'm not talking about the risk of AI, I'm talking about the labor risk, killing, the change of, uh, you know, change of how we're going to do work. The shift is foundational. And if you see our parents generation and the parents before that, such foundational shift happened over a generation, so they could pretty much get into a job, retire. And this has become accelerated now. I think if you see our generation, we're going to look at two to three professions in our lifetime. We're not going to have a single career. And if you see the next wave that is coming, it's going to get more accelerated. This needs a new way of learning, teaching, learn unlearn skill, reskill. This will be the modus operandi. So I'm not seeing a lot of the government, a lot of the educational institutions taking this very seriously. I think when, when this arrives in five years, there'll be a lot of chaos and, you know, a lot of reactive measures, then proactive measures that are taken. And this is a serious conversation, serious conversation that every major entity and government entity should be having at the moment.

‍

Rajan: So, last question, Arshadda. Uh, what is one, uh, book, one tool, and one habit that you would recommend for people that want to actually keep up with the shift that you said that they are all underprepared for?

‍

Arshad Hisham: Okay, so the book would be, uh, yeah, the book would be the Muslim Sullivan's book, you know, that he wrote recently, you know, and that is a very, uh, very good read. And I think, you know, he wrote on DeepMind, and that's one he wrote recently. So. And it's called the coming wave.

‍

Rajan: The coming wave of AI.

‍

Arshad Hisham: Yeah, the coming Fai. And it's aptly titled, that book alone is one that will kind of prepare us what's coming. And in terms of tool, again, it's a bit of a, uh, off channel topic. The tool I would recommend is, again, kind of a book. There are a lot of tools and processes. If you read the old Stoic philosophies, they have a lot of toolkits that prepare you for change in, you know, lot of uncertain times. That's a good one. Yeah. So you, you should use a lot of those frameworks to apply some of the changes coming in, how to retrain, how to reskill, how to enjoy the process, rather than worry about the outcome. So this is kind of some of the key messages, so, some of the tools and frameworks used, and this is kind of very nicely connecting to ancient greenhouse and Rome and the current AI. But I think those tools and frameworks are still relevant to all the things you're doing. Right. I'm sorry, and what was your third one?

‍

Rajan: Wanted to habit. One habit. One new habit.

‍

Arshad Hisham: Yeah, the new habit would, you know, this is very obvious, is learn and unlearn repeatedly learn. And that does not mean to read endless amount of books, just learn and unlearn however you want to learn by watching videos, by doing it. And that's kind of key. And that keeps our, uh, brains young, that keeps us, you know, adaptable to new situation, and we need to be adaptable to the changing circumstance we're going to be. So learn and unlearn is a very important habit that you need to master.

‍

Rajan: Arshad, thank you so much for joining, uh, us. The pivotal clarity show. Uh, it was such a fun conversation with you.

‍

Arshad Hisham: Likewise, Rajan. Enjoyed it. Uh, great questions. Enjoy, uh, the conversation.

‍

Rajan: Thank you.

‍