The Brilliant Boy Billionaire

PART TEN — Franklin

Chapter 5: Settling In

True to my word, I picked the Walters clan up promptly at 6:45 the next evening. The Mach-E was branded as a Ford Mustang, but it was an SUV that didn’t resemble the original Mustang in any way other than the logo. Not only did the Mach-E have significantly more room inside than my Tesla, but it had smoother delivery of power and handled better overall. Although the Tesla Model Y, which was called an SUV, had an extended range per charge of over 300 miles, the all-wheel-drive, extended range performance Mustang Mach-E had a rated range of 270 miles, which was pretty damn good for a much larger car.

The Walters clan fit comfortably in back, and Franklin asked questions about the car during the entire ride, short though it was. Once again, it was Valarie who was waiting for us inside. She took us to the same private room that I’d been in before. Since there hadn’t been time to plan a menu or prepare a buffet, we simply ordered from the menu of the main dining room. The food was as excellent as I’d remembered it.

I had the seafood bisque followed by blackened salmon with asparagus, creamed spinach and garlic-parmesan tater tots. Henry had a Caesar salad followed by a lobster tail, a petit filet, a baked potato and seasonal vegetables. Franklin chose the seafood bisque followed by sea bass over garlic mashed potatoes on a bed of roasted bell peppers and garnished with a roasted-garlic glaze that smelled heavenly. He gave each of us a taste, and heavenly turned out to be an understatement. James and Mora shared a terrine of Marseille bouillabaisse followed by Chateaubriand for two, with creamed spinach and pommes de terre frites, deep fried in truffle oil. I think we all ordered substantial meals in our nervousness and, as a result, were beyond stuffed when we finished them.

As we ate, James continued his story. “Staying in Cincinnati for a few years turned out to be the best thing that could have happened in terms of my career. Not only did I make partner, but I proved to have a real knack for litigating large corporate cases, and that brought in major clients. Not that we gave up on finding you, J.J., but when I was asked to fill the void left in the New York office when one of the senior partners retired, we couldn’t say no. We had to move on for Franklin’s sake and for ours.”

“Believe me, I understand that, considering all the different directions in which Applazon has sent me,” I responded. “You absolutely did the right thing.”

“I realize that now, and I supposed I did back then,” James continued, “but we couldn’t help but feel we were abandoning finding you. It took a long time to get over the feeling of guilt. In any case, I took over a Fortune 500 client list that took my predecessor decades to cultivate.”

“That sounds like a recipe for intense jealousy among the junior partners,” I responded.

“Like you wouldn’t believe,” James confirmed. “Most of them had been vying for the promotion for years only to have a young upstart from nowheresville sweep in and get it. Then to make matters worse, rather than dividing up the top clients among them as they’d been expecting, I took them all, not that it was my intent. Clients asked for me specifically, and the money rolled in. No one ever seemed to complain about that. It took years for the other partners to acknowledge that I was their best litigator and a rainmaker, bringing in enough clients to generate revenue for all. In the meantime, the constant backstabbing took a toll.

“Eventually I couldn’t stand it anymore. The situation was unsustainable to begin with, and it only got worse with time. I made some inquiries and was offered an endowed chair at New York Law School. My department consists only of myself and three other corporate attorneys, but the chairmanship offers the cachet to attract high-profile clients on the side. I negotiated with the firm an amicable divorce without the usual non-compete clause, took a few of the smaller clients with me and have since built a thriving practice based on high-stakes litigation. I only get paid if I win, but I usually win,” he said with a grin.

“And Mora, I know you’re a cardiologist at NYU, but what exactly do you do?” I asked.

“Cardiology is increasing procedure-oriented, and I really liked doing procedures that once could only be done by open-heart surgery. During my fellowship in Cleveland, I did my research on noninvasive valve repair. The field’s so specialized that people who go into it usually do a subspecialty fellowship for an extra year or two, but Cleveland does so many of them, and I did my research in the field, so I had published papers already when I started my work in Cincinnati. Because they didn’t have anyone doing them, they put me in charge of their new noninvasive-cardiothoracic-surgery program.”

“Is that where you do, like, valvuloplasties and tissue-valve replacements by way of cardiac cath?” I asked.

“You’re right on the money, J.J. That’s exactly what I do,” Mora replied.

“She’s actually considered one of the top people in the field worldwide,” James added with obvious pride.

“I have a friend whose father died from severe Covid-19,” I continued. “He had an aortic-valve replacement in his youth for HCM, and I guess that’s a major risk factor for the coronavirus. He nearly died from sudden death as a youth while running cross-country, but one of his teammates ran back to the school and grabbed the AED in time to save him. I read that a lot of those cases are now being treated noninvasively with aortic-tract reconstruction and valvuloplasty. Do you do any of that?”

“I do a lot of those procedures, J.J.,” Mora replied. “It’s one of the main things I do, and I’ve developed a reputation that has brought patients from all over the Northeast and even further. You seem to know a lot about my field. Do you have a particular interest?”

Laughing, Henry replied, “He has a particular interest in everything. He reads everything, and whatever he reads, he remembers.”

“That sounds like a very useful skill,” James responded. “I take it that’s not the same as a photographic memory?”

“The difference is that a person with a photographic memory can remember what’s on the printed page. They can literally see it, and many of them can regurgitate it from memory. I don’t have that skill. I remember only the meaning of what I read. For example, after reading a paper on case law, a person with photographic memory can recite the case and the decision verbatim, even if they don’t understand it. I can tell you the significance of the case and the ramifications of the decision, even though I can’t recite the decision itself.”

“That’s an incredibly useful ability, J.J.,” James acknowledged.

We all groaned when the server brought us our dessert, as we were already beyond stuffed. We’d ordered bourbon bananas foster for the table, which she flamed before serving each of us. As wonderful as it smelled, we couldn’t but help it when we picked up our forks. As we ate our dessert, Franklin asked about my interest in auto racing, so I filled him in on what I’d been doing.

When I told him the specifics of the problem we’d been having with waste-heat generation, Franklin responded, “You’re making things so much more difficult than they need to be. Physicists do everything the hard way, using differential equations when a Fourier, Laplace or Z transform quickly reduces the solution to simple algebra.

“Look, you’re probably familiar with Ohm’s Law and terms like resistance, reactance and impedance, but the moment you introduce capacitive and inductive reactance, you end up with differential equations and a real mess. In electrical engineering, we take a Fourier Transform of the equations to reduce the differential terms to simple algebraic expressions. Reactance simply becomes the imaginary component of impedance, with resistance being the real part. It lets you combine the two and makes solving the system of equations quite easy.”

“You use complex numbers to represent the amplitude and phase of a periodic signal,” Henry translated. “I get that the two approaches are mathematically equivalent, but how does that help us?”

“You’re using an alternating electric current to generate a sinusoidal magnetic field on a super­conducting disc, with a rotating phase component that drags the rotor with it, as per the Hall effect. The ceramic substrate isn’t really a super­conductor though ’cause the current flows through it by quantum tunneling; am I right?”

“It took an untold number of experiments to figure that out,” I confirmed, “but once we figured out why it works, we were able to devise ceramics that didn’t rely on rare-earth elements for magnetism and yet to extend the effect to room temperatures and above.”

“Although quantum tunneling acts like super­conductivity, there’s one important difference,” Franklin continued. “Resistance may be zero, but you still have reactance and hence impedance. There’s a small component of capacitance that may become significant when there are sudden transients, and of course there’s induction from the magnetic fields, which is at its strongest at low frequencies and thus low speeds — or even standing still.”

Instantly, I saw what Franklin was getting at. It was pretty obvious, but it took Franklin, with his studies of electrical engineering in his classes at HSMSE, to recognize it in its most basic form. “The question is how we eliminate the reactive loss,” I challenged.

“You can’t eliminate it entirely,” Franklin responded, “but you can mitigate it. For example, capacitance is inversely proportional to distance, so the thicker you make your ceramic elements, the lower the capacitance will be. Because the tunneling effect isn’t dependent on distance, a thicker stator won’t affect super­conductivity.”

“But it will affect magnetic flux,” I pointed out, “particularly with the Hall effect.”

“I’m not exactly sure how to model it, but with rotating magnetic fields and a rotating disc, the magnetic field has a sort of angular momentum,” Franklin continued. “That, more than anything, is responsible for the reactive loss and heat generation. You could virtually eliminate the effect by using interleaved concentric cylinders instead of discs…”

“At the expense of losing much of the power from the Hall effect,” I countered.

“Why are you so wedded to the Hall effect?” Franklin asked.

“For one thing, I’d like to avoid the use of rare-earth elements for the strong, permanent magnets used in most motors,” I explained. “They’re mined in the Third World by what amounts to slave labor, often by children who can get into the tightest of spaces. Secondly, I’d like to avoid the use of interconnects. The Peltier cooling effect in the rotor is powered by a battery in the hub of the disc that charges through magnetic induction during regenerative braking, but you’d need much more power than could be supplied by a local battery for electromagnetism. Finally, the use of the Hall effect is really cool, and it’s my idea, so I guess it’s a bit of an ego thing.”

“I get that you don’t want to rely on niobium for permanent magnets, even though everyone else does,” Franklin responded. “Eliminating its use would be a game changer and would make green energy much greener, particularly when it comes to wind turbines…”

“Actually, I’ve already designed a wind-powered generator based on the Seebeck effect,” I interrupted. “Wind passing around an obstruction, such as a cylinder, causes a temperature differential between the front and back surfaces that we can use to generate an electric current. We’re building a prototype wind farm in Eastern Washington and at least on paper, we should be able to realize power outputs comparable to that of a conventional wind turbine without the need for any moving parts. We haven’t even started to explore optimization, so I expect there are major gains to be made from optimizing the geometry, if nothing else.”

“That’s really cool,” Franklin exclaimed. “Getting back to your motor, however, it’s not worth saving your ego if it doesn’t work.” That was a slap in the face, but probably a needed one. It was the first time I realized that I had an ego, much less that I’d let it get in the way of my judgment.

“Okay, let’s see if we can figure out how to make a ceramic permanent magnet,” I suggested.

“Maybe not a permanent magnet, but a super­conducting magnetic channel,” Henry suggested.

The moment Henry started to speak, I knew exactly how we could do it. As Franklin had suggested, the stator and rotor could consist of interleaved concentric cylinders, with the rotor attached to the wheel and the stator to the axle, each within its own housing. If the cylinders of the stator were super­conducting in the axial direction and the rotor in the radial direction, the resulting magnetic field would bond the two together.

By rotating the phase of the magnetic fields in the stator, the rotor would physically rotate with the fields in perfect lockstep, without the reactive losses associated with the current design. Transients from irregularities in the road surface wouldn’t even be an issue, and regenerative braking would be built-in.

Poor Mora and James took it all in stride as the three of us teenagers worked on the new design for the motor. I used my phone to send a preliminary design to Jitendra, who immediately acknowledged receiving it. Because it was a radically different design from the prototypes we’d been building, I knew it would take months to build and test a new prototype, but I was optimistic that we were on the right track.

<> <> <>

After Jerry’s funeral and the remarkable revelation that Mora and James were my parents and Franklin was my brother, we were in no mood to return to our vacation. Undoubtedly, we’d plan a trip for sightseeing and hiking in Canada sometime in the future. We stayed in Omaha for a couple more weeks, helping the family with the transition to life without the man who’d been such a powerful presence in the home. However, Fran was not a wallflower by any means. She’d always been strong-willed, and that didn’t change in the absence of her husband.

Having bought the New York condo furnished, moving in amounted to little more than carrying our luggage in the front door. The home theater was finished, and the equipment for our home gym was on order and would be installed within the next couple of weeks. The sellers elected not to keep anything from the apartment, so we sold all the artwork, netting a princely $28 thousand for a thoroughly mediocre collection that was appraised at double that and for which the sellers undoubtedly paid even more. We tried donating the collection to several area hospitals, only to be told they couldn’t use it. It just wasn’t worth the trouble to even give it away, so we offered it to the appraiser for the appraised value, and she said she’d take it off our hands for half that. We readily agreed.

All of the bedding needed to be replaced, so we went online and read everything we could about recommended mattresses. There sure were a lot of mattress brands to choose from, and not surprisingly, it seemed that, with the exception of Casper, some of the ones most heavily advertised on TV were among the most poorly rated. We tried out a lot of mattresses and took advantage of offers to try them out in our home before settling on our first choice. We liked the Casper and even tried the top-of-the-line Wave, but the Wave went back. Nothing compared to the top-rated Avocado Green, so we eschewed the ever-popular memory foam options in favor of a mid-priced, eco-friendly innerspring mattress. We then went shopping for and purchased new bedding for each of the three bedrooms and towels for the bathrooms. We could’ve bought it from Applazon, but designer bedding was not one of Applazon’s strengths.

We took advantage of my company discount to fully stock the kitchen at the Applazon Organic Market. A stop at the nearby Container Store gave us ideas for implements for the kitchen, and a trip to Best Buy gave us ideas for updates to our small appliances. We ordered everything from Applazon, of course. That still left the artwork and little knickknacks to obtain before the place could truly be our own. Our next-door neighbors, Max and Gideon, were planning a trip with us to the nearby Rubin Museum once we were settled in. Further, they promised to introduce us to a variety of artists who could furnish our apartment with unique artwork that suited our tastes.

No sooner did we settle in than it was time to register for classes at Columbia. I needed to schedule a proctored qualifying exam that covered a master’s level curriculum in machine learning and A.I. before I’d be permitted to begin work on my Ph.D. A computer-science undergraduate degree was acceptable in meeting the prerequisites, but the core graduate curriculum in A.I. was essential to any graduate degree in the field. Unlike the computer-science QE I’d already completed, this one wasn’t open book, nor could it be completed at home. An open-book, take-home exam by its nature requires a lot of thought, with questions that would challenge a professional.

The notion that Columbia used a proctored QE to make it easier to write questions was quickly dispelled the moment I opened the test booklet. The questions involved solving real-world problems, but within the span of a four-hour exam and without access to outside resources. I doubt that I could have passed it had I not spent time reading the literature and were it not for my memory abilities. Every answer not only needed to be a viable solution but had to be justified based on the use of specific, named algorithms or citations of literature by author and year. It was one of the most challenging exams I’d ever taken and the first exam I couldn’t say I aced. I received a passing score, but several of the students taking it did better than I, and I came to appreciate one of the key advantages of attending an Ivy League school: the intellect of the students. For the first time in my life, I’d be with students who could challenge me.

My initial disappointment that I couldn’t simply read the material and take the exams was replaced by a profound appreciation for my peers, all of whom were graduate students working on their Ph.D. degrees in aspects of machine learning and artificial intelligence. There were only about a dozen of us, but the fact that one institution could attract that many exceptional young people was amazing.

I wasn’t the only teenager in the group either, and Robin and I quickly gravitated toward each other. Like Henry, she was still fifteen, turning sixteen in mid-October, yet she already had a slew of publications to her name as well as an algorithm named after her. We had very different approaches to solving problems and would spend hours outside of class arguing with each other by text. Our disagreements in class were truly epic, which was why my initial resentment at having to schlep my way uptown was soon replaced by my looking forward to the twice-weekly seminar format. I came to appreciate our personal interactions. Her abilities complemented mine, which was why I became bound and determined to recruit her to my division at Applazon.

That it would be an uphill battle was obvious from the beginning, so I settled into an effort I expected would take the rest of the academic year. If I didn’t succeed by then, I’d have to recognize the futility of the effort. The trouble was that, to me, failure was a sort of vindication of the man I’d thought was my father. Needless to say, I was loath to accept defeat. Unfortunately for me, Robin was herself bound and determined to pursue an academic career. She hated the corporate culture and everything associated with it and was dead set against lining the pockets of corporate fat cats like Jeff — or me. From my perspective, Applazon had been incredible, supporting me in all my endeavors. To Robin, they were the evil empire, and Jeff was personally responsible for the death of brick-and-mortar retail. Where I saw the natural evolution of the marketplace, she saw vacant storefronts and a business model that fostered global warming.

Then again, I liked a good challenge.

<> <> <>

“May I come in?” Larry asked as he approached my desk.

“You know my door’s always open for you,” I replied with a laugh as I moved to the small conference table next to my desk and motioned for him to sit down with me. Larry entered my doorless cubicle and sat down across from me. Then thinking I should be a better host, I asked, “Could I get you a cup of coffee or tea or something?” I asked. “I have some organic hummus and a variety of organic cheeses with multigrain crackers. I also have a selection of Greek yoghurts.”

“I’d love some Earl Grey tea, if you have it,” Larry responded. “Thanks.” I located an Earl Grey K-cup and placed it in the Keurig machine, brewing a mug of tea for my guest. After handing him his tea, I brewed a mug of Dead Man’s Reach coffee for myself. I wasn’t a particular fan of K-cups for a variety of reasons, not least of which was the environmental waste, but I’d found that guests didn’t appreciate waiting around for me to grind and brew fresh coffee while trying to talk to me. With the Keurig, I only needed to insert the appropriate K-cup into the machine. I was thrilled that Raven’s Brew now sold K-cup versions of my favorite coffee.

Sitting back down at the table with Larry Cohen, the deputy director of the New York division of A.I., I looked out at the mostly empty workstations where the other members of our division would soon go about their work. We were still in temporary space, as it would take more than a year for Applazon to gut and remodel the eighth floor, let alone build out our little section of it. Had it not been for Larry’s and my exploration of it, the eighth floor might well have languished for years, being developed only after all the other space in the building had been put to use. However, once it was known that there was so much undeveloped raw space to be had, ours was but the first of an avalanche of requests for space on the eighth floor, and corporate had little choice but to reevaluate their plans. Still, management was very tight-lipped about who was being considered for space on the eighth floor.

By being first, we were able to secure the prime location I’d requested, right at the corner on Fifth Avenue, and it would be developed exactly as I wanted it, with a two-story atrium. Unfortunately, it wouldn’t be ready until the summer of 2024. In the meantime, we’d have to make do with our temporary location in a large open office with cubicles and without any windows. It was no worse than what we would have gotten had we not found the eighth floor, and I could live with that until we moved into our permanent space. As yet, we were the only division with a firm commitment to the eighth floor. I was grateful for that.

Larry had already laid much of the groundwork for the assembly of a team of professionals. The resumés I’d been reviewing were all exceptional, making it difficult for me to decide whom to recommend he invite for interviews in the next round of recruitments. As I’d already agreed in accepting the job, I would honor Larry’s decisions on the makeup of the division, but that didn’t mean I wouldn’t play a significant role in this decision-making process. As we got to know each other, he was deferring more and more to my judgment. He was more than twice my age — my real age — but he accepted me as his peer. His willingness to defer to a mere teenager went far in my book, but perhaps my willingness to defer to his experience had just as much to do with how well we got along.

“So, what’s up?” I asked Larry.

“I came across something interesting,” he said as he handed me a reprint of a paper by Frank Drüsen and Richard Katz.

“Paper. How quaint,” I said as I took the proffered manuscript from Larry and started to read.

“This one’s worth cogitating on,” Larry explained, “and sometimes it’s just easier with a paper manuscript.”

The title was intriguing enough, so much so that I read it aloud, “The Cognitive Lattice: a novel approach to machine learning.” As I read the abstract, it quickly became apparent that they were onto something. In contrast to classical mathematics, which uses probability theory and statistics to correlate cause with effect, traditional machine learning doesn’t assume that such relationships are linear or even consistent. Utilizing very large data sets and numeric methods, they attempt to mimic the natural process of human learning in which causal relationships are inferred from observation.

For example, a statistical approach to facial recognition might assume there are measurable features such as eye spacing and cheekbone height that can be correlated with individuals in a data set. Machine learning algorithms assume nothing. Presented with thousands of photos of people, the algorithms ‘learn’ what differentiates one person from all the others. In effect, the computer becomes a ‘black box’ in which photos go in and identities came out, without anyone really knowing exactly how the computer applies its logic.

Problems have arisen when the algorithms are applied to people from outside the data set, particularly if they don’t resemble those on which the algorithm was trained. In the majority of cases, the data set on which they were based included a preponderance of white men, hence they’ve performed poorly in analyzing photos of people who aren’t white men and particularly with women of color. People have inherent biases, too, but at least they’re better able to reject making false matches than are machines, which are programmed to identify relationships, regardless. Unfortunately, a lot of innocent individuals have been and continue to be apprehended, based on the assumption that the algorithms are always right.

One of the worst examples of A.I. failure came from Applazon, in which a school district used our software package as the sole means for making decisions on hiring, firing and awarding tenure. In one particularly egregious case, a teacher with numerous teaching awards was deemed subpar by the software and denied tenure. The lawsuit that resulted cost Applazon millions and did much to tarnish its reputation. It was because of that debacle that Jeff decided to start over with a new group of researchers from outside the Seattle culture. I was determined to make use of the massively parallel architecture of our new data mini-centers to explore novel approaches to A.I. The approach described in the paper by Drüsen and Katz was precisely the sort of thing I was looking for.

Richard Katz apparently was a post-doctoral fellow at Carnegie Mellon, and Frank Drüsen was one of the doctoral candidates working with him. The footnotes indicated that the paper was based on the results of Drüsen’s dissertation, which meant that he was likely in his own post-doc by now or had been hired by industry. Katz had probably moved into a junior-faculty position by now, either at Carnegie Mellon or elsewhere. In any case, the paper described a mesh-based approach to machine learning more akin to solving a set of simultaneous equations. For each input variable, every possible outcome was explored, building a mesh table of probabilities for all variables in the data set. The method was particularly amenable to solution by quantum computing because of its massive scale and its ability to look for patterns in seemingly unrelated data.

“I’m going to have to get a copy of Drüsen’s dissertation,” I said, at which point Larry literally tossed a copy of it in front of me. Skimming through it quickly, I ascertained right away its applicability to our work. “This would be easy to implement on our quantum-computing platform. I wonder if they’re both still in Pittsburgh.”

“It seems they’re a couple,” Larry replied. “Katz is an assistant professor now, and Drüsen is a post-doc in his lab. Would you like to interview them? We’d have to offer them both jobs or it would be a no-go for them for sure.”

“Drüsen’s dissertation catapults them to near the top of the pile, I think. Don’t you?” I asked.

“If they’re interested and if they interview well, I’d like them here with us,” Larry agreed. “They’re too good for a place like Carnegie-Mellon. They’ll languish there.”

“I couldn’t agree more,” I chimed in. “We can probably get them on the adjunct faculty at Columbia. At least they’ll be appreciated there.”

“Agreed,” Larry replied. “Would you like to be in on the video interviews?”

“I’d like to meet them in person,” I countered. “I think I can authorize it if Grace signs off. I hear that Pittsburgh is rather nice in the fall, also. Maybe we can go there for the interviews.”

“I’ve never been there, either,” Larry responded. “Let’s go.”

“It’ll have to wait until after Henry’s birthday, though,” I interjected.

“He’ll be what, sixteen?”

“Yeah,” I replied, “it’ll a major milestone for my baby.”

“Word on the street is that you’re only sixteen,” Larry revealed. I wasn’t at all surprised that cracks in my carefully crafted identity were already beginning to appear, particularly within the organization. Too many people knew as it was.

“The rumors are true, but that’s not for public knowledge — at least not yet,” I explained. “I’ll actually be seventeen two days before Thanksgiving. The corporate types are working on a strategy to address the inevitability of the truth coming out once the media starts prying into the backstory behind the kid they’re already calling the ‘brilliant boy billionaire’.”

“Brilliant boy billionaire?” Larry asked with a bemused expression.

“I think Shira Ovide coined the expression. She’s a technology writer at the New York Times.”

“I know who Shira Ovide is,” Larry said. “She’s interviewed me a couple of times, and you can expect the same from her and then some. So, we’ll go to Pittsburgh after Henry’s birthday?”

“Definitely,” I agreed. “See what you can set up.”

Unfortunately, we never got the chance.