Part IThe Story
The Witness Who Debugged Himself
The question had been asked once before, under oath, on May 25, 2007, by investigators from the Securities and Exchange Commission. It came back in a federal courtroom in Santa Ana in the second week of December 2009, in a slightly different form, from the mouth of a defense lawyer. That lawyer represented William J. Ruehle, Broadcom's former chief financial officer, who was on trial for securities fraud. The witness was the man who had co-founded the company, a 55-year-old electrical engineer with a soft voice, a professor's habit of qualifying every sentence, and a felony guilty plea on his record. He was testifying under a grant of immunity. He had not yet been sentenced.
"Do you recall whether that statement involved whether or not you were involved in granting options to officers of the company, which was the job of the compensation committee, and that you indicated that you were not so involved?"
Henry Samueli did what he had been trained to do since he was twelve years old and holding a soldering iron. He took the failure apart.
"I recall the question was about my involvement in the granting process of options to Section 16 officers, who are the highest executives of the company, and… that is the purview of the compensation committee. And I replied that I wasn't involved in the granting process, which was not truthful because, in fact, I was involved in making recommendations to the compensation committee." Then the engineer's frustration, which is a particular kind of frustration — not at being caught but at having produced an output that the inputs didn't justify. "But ironically, the thing that bothers me most about that is that there is nothing wrong with being involved in the process of making grants to Section 16 officers. I'm perfectly allowed to make recommendations. So, why I would say I wasn't involved escapes me."
He kept going. He had thought "long and hard about it because that's not my style." His best hypothesis was a parsing error: he had heard compensation committee, known he was not on it, and answered the question he expected rather than the question asked. "I made that false statement and I accept it, but I bang myself in the head for making such a stupid statement."
It is possible to read that testimony as a well-coached performance. It is also possible to read it as exactly what it sounds like — a perfectionist conducting a root-cause analysis on his own mouth in front of a jury. U.S. District Judge Cormac J. Carney, who fifteen months earlier had refused to let Samueli walk away with probation and a $12 million check because it would look like "justice is for sale," listened to two days of it. Then he asked the witness to step down and face the bench.
What happened next silenced a packed courtroom, and it will be returned to. For now, it is enough to hold the image: the most successful communications engineer of his generation, the man whose chip designs put broadband into tens of millions of American homes, standing in front of a federal judge, trying to explain why a signal had come out garbled. He had spent his whole life on one question — how does the sound get out of the box — and he was being asked, at last, about the one transmission he couldn't account for.
By the Numbers
Henry Samueli
$5,000Each founder's personal stake when Broadcom launched in August 1991
95%+Share of U.S. digital cable set-top box and cable modem chips designed by Broadcom, per IEEE Spectrum (1999)
$2.24BBroadcom's 2007 restatement of options-related expenses — the largest of the backdating era
$37BPrice Avago agreed to pay for Broadcom Corporation on May 28, 2015
75+U.S. patents bearing Samueli's name
$307.7MHenry and Susan Samueli's giving to UC Irvine, its largest donors (OCBJ, 2025)
1995The year his UCLA leave of absence began. He is still listed on the faculty.
A Ninth-Grade Radio in a Seventh-Grade Hand
Sala and Aron Samueli came to the United States from Poland in 1950. They were Holocaust survivors, Jewish, and arrived — the word their son would use for the rest of his life — with "virtually nothing." They landed in Buffalo, New York, where immigrant aid organizations found them housing, food, work. Henry was born there on September 20, 1954. When he was ten, the family moved to Los Angeles, and the parents bought a small liquor store in East Los Angeles. Their son stocked the shelves. He worked the register. He worked weekends there all the way through middle school and high school.
Decades later, sitting on a stage at UC Irvine's business school in November 2019, Samueli would explain the origin of everything that followed in a single, plain causal chain: the organizations that had helped his parents in Buffalo; their gratitude; the obligation they passed down. "They were so grateful for that, that they certainly instilled in me from a young age the importance of giving back to those in need, and those less fortunate than you," he said. "And that's been impounded in me my whole life."
Impounded is an odd word, and a revealing one. Not instilled, not taught. Held, as in a reservoir, under pressure.
The other origin — the one every profile tells, and the one he tells best — happened in a seventh-grade shop class at Hubert Howe Bancroft Junior High in the West Hollywood flats. Shop was required. He chose electric shop. Each student was to build a crystal radio from a kit: a single crystal, an antenna wound around a toilet-paper tube. Samueli found it boring. In a Heathkit catalog he had seen a five-tube shortwave radio, and he asked the teacher if he could build that instead. The teacher said no. The shortwave set was a ninth-grade project. Samueli kept asking until the teacher gave in.
It was not easy. It was, as IEEE Spectrum's Tekla Perry put it in a 1999 profile, a "cookbook project," but there were hundreds of solder joints and he had never done anything like it; he taught himself to solder in the process. "I worked every night," he recalled. "It took a full semester to build it." At the end of the term he carried the box to school. The teacher plugged it into the wall. Sound came out.
"The teacher's jaw hit the floor," Samueli said. "He said nobody gets it right the first time." The teacher told the boy he would be a successful electrical engineer someday. It was the first time Samueli had heard that such a profession existed.
There is a detail here that the authorized versions sometimes blur. The radio wasn't a Heathkit at all. Samueli later went back and determined that it was a Graymark 506B — and that a man worth tens of billions of dollars cared enough about the provenance of a 1960s hobby kit to correct the record tells you something about his relationship to precision. When he endowed a science competition for middle schoolers decades later, the press materials specified the brand. Graymark.
The second detail matters more. Samueli was not surprised the radio worked. "I wasn't surprised it worked — I'm sort of a perfectionist — but I had no idea how it worked." That was the problem. He had followed every instruction correctly and produced a thing he did not understand. "It was magic," he said in 1999. "I was amazed that how, out of thin air, you get sound coming out of a speaker." Then the sentence he has repeated, in nearly identical words, to the Los Angeles Times, to IEEE Spectrum, to the National Inventors Hall of Fame, to auditoriums full of UCLA undergraduates:
— Henry Samueli, to IEEE Spectrum (1999)That became my mission in life, from seventh grade onward, to find out how radios work.
It took nine years of college, a doctoral thesis, and several years in industry before he felt he had satisfied the mission. "By the time I received my Ph.D. in electrical engineering," he said, "I finally did understand it."
He kept the radio. In December 1999, the month he gave away $50 million, he mentioned to two Times reporters that it was stored in his garage.
The Exact Science
He graduated from Fairfax High School at sixteen and enrolled at UCLA in 1971. It was the only college he applied to. Tuition was $600 a year. He lived at home to save money — through the bachelor's degree, completed in 1975, when he was twenty; through the master's in 1976; through the doctorate, finished in 1980, just before his twenty-sixth birthday. All three in electrical engineering. All three at the same campus, a few miles from the liquor store.
"My parents could not have afforded to send me to a private school," he said later, and he said it the way people say things that remain present tense.
At UCLA, Professor Alan N. Willson Jr. taught a new course on digital signal processing, and Samueli, by his own account, "fell in love with it." The explanation he gave a room full of engineering students in January 2018 — with Willson sitting in the audience — is the closest thing on record to a self-portrait:
"I'm a perfectionist-type of person. I just like everything to be exact and perfect and the beauty about digital signal processing is it's exact. You know exactly what the computations are."
Analog signals are a world of approximation, drift, noise, tolerance. Digital signal processing promises that a signal, once converted to numbers, can be manipulated with arithmetic certainty. For a boy who had built a radio that worked without his understanding why, DSP offered the reverse proposition: nothing would happen that he could not account for.
The dissertation Willson supervised was titled "Nonperiodic forced overflow oscillations in digital filters." It is a highly theoretical paper about a specific pathology — the way a digital filter, pushed past the limits of its number representation, can lock into oscillations that have nothing to do with its input. A system producing output that its input doesn't justify. Samueli and Willson published a version in IEEE Transactions on Circuits and Systems in 1983. It is an obscure thing to have spent a doctorate on, and an apt one.
After the Ph.D., in 1980, he went to TRW in Redondo Beach, to the Electronics and Technology Division, which built military broadband communications systems. He called it "a tremendous opportunity to learn because you're dealing with superhigh tech, the greatest technology at the time." He also kept a foot in Westwood, teaching part-time as a visiting lecturer. A.R. Frank Wazzan, the engineering school's dean, watched him teach and decided to break a rule. UCLA did not hire its own graduates onto the faculty. In 1985, Wazzan offered Samueli a full-time professorship anyway.
"I sort of babied him," Wazzan said in 1999. "He was very bright and very promising."
Samueli took what the Los Angeles Times described as a huge pay cut to accept. In 1989, the school gave him the TRW Excellence in Teaching Award — named, fittingly, for the defense contractor he had left. With fellow professor Asad Abidi, he built a research program in UCLA's Integrated Circuits and Systems Laboratory that set out to do something that was not yet commercially obvious: put the signal-processing machinery of high-speed digital communication onto cheap CMOS chips. In 1991, his group won a Meritorious Paper Award at the Government Microcircuit Applications Conference for a paper whose title reads, in retrospect, like a product roadmap: "CMOS Integrated Circuits for High Bit-Rate Digital Modems, Adaptive Equalizers and Frequency Synthesizers."
Somewhere in that same stretch — the sources disagree on the decade, and he has never seemed interested in fixing it — he met Susan at a dance at a Jewish temple in Los Angeles. She was a Valley girl, in the literal sense, born and raised in the San Fernando Valley, a 1972 mathematics graduate of UC Berkeley who had gone to IBM as a software programmer in the Federal Systems Division and moved into systems engineering in the company's National Marketing Division. Two engineers. She left IBM in 1985 to raise their three daughters, Leslie, Jillian and Erin, and, out of a refusal to give the girls antibiotics "every time they had an earache or a cough," as Henry later put it, she began studying nutrition, homeopathy and Chinese herbs — eventually earning a doctorate in nutrition from the American Holistic College of Nutrition in 1993 and a diploma from the British Institute of Homeopathy in 1994. It is worth setting this down early, beside the man who loves DSP because "it's exact." The marriage contains its own unresolved oscillation, and it would later carry a $200 million price tag.
Two Henrys
The student who walked into Samueli's first doctoral advising slot was six foot six.
Henry Thompson Nicholas III was born in Cincinnati on October 8, 1959. His parents divorced when he was four; his mother, a schoolteacher, moved him and his sister to Los Angeles and later married a journalist and Hollywood screenwriter named Robert Leach. Nicholas went to elementary schools in Malibu and to Santa Monica High, graduated in 1977, and enrolled at the U.S. Air Force Academy in Colorado Springs to fly fighter jets — and left when he was judged too tall to fit the cockpit. He had dyslexia and, by his own account, attention deficit disorder; as a child he had trouble reading and doing simple math. He finished an electrical engineering degree at UCLA in 1982 and went to work at TRW in Redondo Beach, where he met a young engineer five years his senior named Henry Samueli. In 1983, his sister Marsy was murdered. He would spend the rest of his life, and a great deal of his fortune, on the rights of crime victims.
The early friendship ran in one direction. Samueli provided; Nicholas followed. When Samueli left TRW for the UCLA faculty in 1985, he persuaded Nicholas to come with him as his first Ph.D. student. In 1988, when Samueli took a part-time consulting role at PairGain Technologies in Cerritos — he would become its chief scientist — he talked Nicholas into joining to run an integrated-circuits team. The two of them built PairGain's first high-speed digital subscriber line transceiver, the box that let businesses make digital connections to their local phone companies. It shipped in 1991 and took 80 percent of its market.
That same year, the direction reversed. Nicholas left PairGain and, in August 1991, opened a company in the spare bedroom of his condominium. Each man put in $5,000 of his own money. The company was first called Broadband Telecom; its logo, designed by Nicholas's then-wife Stacey — herself a former electrical engineer — was inspired by the sinc function, the mathematical curve that describes how an ideal low-pass filter responds to an impulse. A signal-processing joke as a corporate seal. It survived, redesigned, into the Broadcom logo.
Samueli did not quit his job. He stayed on the UCLA faculty and kept consulting for PairGain through 1994. He hired his graduate students as part-time employees. It was Nicholas who, in 1995, finally talked him into leaving — the student recruiting the professor, the inverse of every move that had come before. Samueli asked Wazzan for a leave of absence. Wazzan, making his second exception for the young man he had babied, granted it indefinitely.
The leave began in 1995. It has not ended. On the UCLA electrical and computer engineering website, Henry Samueli is still listed as Professor, Circuits and Embedded Systems, Office 58-121 Engineering IV, with a phone number and an email address.
What the two Henrys became, in the decade that followed, is best described by the people who watched them. "Henry Nicholas was the fire in that enterprise and the other Henry was more deliberate," said Anil Puri, the business school dean at Cal State Fullerton, "and the combination of the two was exactly what the company needed." The Times, in 2015, put it more starkly: "one man deliberate and reserved, the other anything but." When they got rich, the reserved Henry bought a house in Corona del Mar, toys for his three daughters, and a Porsche Boxster — for one of his employees. He kept driving his leased Cadillac. The brash Henry bought a Harley-Davidson, a red Ferrari, a black Lamborghini, a 15,000-square-foot mansion in Laguna Hills with waterfalls and two pools, and, for longer trips, a Las Vegas charter aviation service with three jets and a helicopter. "I get what I want," Nicholas told a reporter in 2000, "because I always want to win."
Forbes, the same year, noted that Nicholas's assistant scheduled sleep into his PalmPilot so he wouldn't forget to take it.
The partnership has the shape of a stock literary device — the angel and the engine, Apollo and Dionysus, Dr. Jekyll and the man Bethany McLean would title her 2008 Vanity Fair profile after. But the most precise description came from Nicholas himself, in 2001, and it is not flattering to either of them:
— Henry T. Nicholas III, to the Los Angeles Times (2001)The degree to which I push some of the engineers, Henry would never come close to that. But I would never be able to push them as hard as I do if I didn't have Henry on the positive side.
The two halves of a push-pull amplifier. Neither one does anything useful alone. Each one exists because the other one does.
Pulled, Not Pushed
Broadcom, Samueli liked to say, never had a business plan that a venture capitalist would recognize. It never needed one. It never took venture money at all.
The first revenue came from where Samueli had come from: small defense contracts, the kind of work TRW had trained both founders to do. "These projects funded our R&D, and we gained more and more knowledge," he said. In a 2009 talk at UC Irvine, he credited the military customers with something more than cash. "We ended up with a good two-year lead over the competition, and that's what launched us into being a major, successful company." The company never spent beyond the terms of its contracts. Growth was slow, careful, and paid for.
Then the first major commercial customer: Scientific-Atlanta, the cable equipment maker, which needed a chipset for what would become the world's first commercially deployed digital cable television receiver. Broadcom delivered it, and then reengineered the chipset into a single chip. And then Samueli saw the larger thing. The same system-on-a-chip architecture that could pull digital television off a coaxial cable could pull data off it too. Cable modems. In the dial-up era, when the shriek of a 56k handshake was the sound of going online, Broadcom's single-chip cable modem promised, as PCMag later summarized it, 50 megabits per second — on the order of a thousand times faster.
"We didn't push the technology out there," Samueli told the National Inventors Hall of Fame. "It was pulled out of us at the university because the demand was so great in the industry for a fast digital" — and the sentence trails into the institutional record there, which is perhaps the right place for it to trail.
The company moved from the bedroom to a small office in Westwood in 1992, near enough to campus that graduate students could walk over, and in 1995 to the Irvine Spectrum in Orange County. Samueli went full time. Three years later, in April 1998, Broadcom went public on Nasdaq under the ticker BRCM. Nicholas, characteristically, turned away some institutional fund managers who wanted allocations because he believed they didn't understand the business. "We've managed to break all the rules," he told Forbes.
The rules they broke had a structure. Forbes laid it out in June 2000: pick the largest customer in each networking-chip market; design the best chip for that customer; build it cheaper than the incumbent supplier. Take the account away from Lucent, from Intel. And — the audacious part — get order commitments before building a prototype. "We show them the specs and convince them that we can build something they thought was impossible to build," Nicholas said. A company of Ph.D.s, staffed with a stated hiring preference for academics, selling paper to engineers on the strength of math.
By September 1999, IEEE Spectrum reported that Broadcom made more than 95 percent of the chips going into U.S. digital cable set-top boxes and cable modems. By 2000, according to the National Inventors Hall of Fame, its chips were in 90 percent of cable modems and set-top boxes and in more than 60 percent of all local-area networks. Samueli's patents covered DSP-based receiver architectures for cable television, satellite television, Ethernet, and high-bit-rate DSL. The one the Hall of Fame chose to cite, decades later, was U.S. Patent No. 5,754,591 — a method for processing quadrature-amplitude-modulated signals, the encoding scheme by which a single carrier wave is made to carry many bits at once. If you are asked which of his inventions he loves most, he will say the cable modem patents. "I have a really soft spot in my heart for them."
He has always been unusually candid about luck. At a 2014 lecture at UC Irvine — he gives some version of the talk there most winters, titled "The Story of Broadcom: How a UCLA Professor Became a Successful Entrepreneur" — he put up a slide and played the screeching dial-up tones for comic effect, then told the students the thing founders rarely say aloud. "Timing is everything. If Broadcom arrived ten years earlier, no way we would've been successful, even five years later we never would've been successful either." In 2009: "To be successful, it takes a lot of luck. There are things that are totally out of your control."
The radio had worked the first time. He had not been surprised. He had simply not understood why. Broadcom worked too, and the founder's most consistent explanation of its success is a window in time that he did not choose.
The Calm Father in a Loud House
In August 2000, Broadcom's stock traded at nearly $275 a share. Forbes had just put Nicholas at 65th on its list of the world's richest people, $5.4 billion, and Samueli at 70th, $5.1 billion. The company expected annualized sales of $1 billion by the end of the year. It had been profitable since inception.
Asked about the money that June, both men performed the indifference expected of ex-geeks. "I couldn't care less whether I have $10 million or $4 billion," Nicholas said. "Either way, I'll spend the same amount of money between now and when I die." Samueli, in a separate interview: "The billionaire thing doesn't register. I don't have any time to spend it."
What they did spend, mostly, was stock. Between January 1999 and January 2001, Broadcom bought company after company with its own paper: Epigram for $316 million, HotHaus for $280 million, Innovent for $440 million, Altima for $533 million, Silicon Spice for $1.25 billion, Newport Communications for $1.32 billion, SiByte for $2.06 billion, ServerWorks for roughly a billion. Eighteen acquisitions in less than two years. Samueli spent much of that time, the Times reported, "trolling companies worldwide for cutting-edge technology and talented engineers." Employees were showered with stock options that made hundreds of them millionaires. The parking lot in Irvine, the Times would later write, gleamed with Porsches and Lamborghinis while engineers worked late into the night. Samueli would testify that some of them worked 80-hour weeks preparing for the IPO.
Then the bubble broke. By June 2001 the stock was in the $30 range. On a morning that month, with the shades drawn in a conference room at the Irvine headquarters, Samueli gathered two dozen of his top engineers in front of a slide screen that projected, as the Times reporter Karen Alexander wrote, "the grim mathematics of a technology market slowdown." The engineers offered ideas. Share costly test equipment. Forgo salaries in exchange for more stock. The taciturn chief technologist's voice took on an uncharacteristic edge.
"You can do back flips, and you're still not going to salvage" the quarter, he told them. Layoffs were coming. "We have to do it and get it out of the way and start with a clean slate. We can't keep talking about negative things. We have to start getting morale up."
Start with a clean slate. It is the kind of phrase that sounds like nothing when it's said — a manager's phrase, a way of closing a meeting. Hold onto it.
This was the moment, the Times observed, when "the spotlight has turned to the brainy Samueli." In public, he had been "a nearly invisible counterpoint" to Nicholas, the brash symbol of the new economy. In private, he had been something else: "a calming father figure to his beloved engineers. When Nicholas made his overworked engineering staff cry, it was Samueli who rebuilt their confidence and set them back on task." He led more than 1,900 engineers. His decisions about which projects to kill and which to fund would determine, as one of his senior engineering directors put it, whether Broadcom could see "multiple years in the future."
There is something worth noticing in the arithmetic of the story Samueli would later tell a jury. Stock options that couldn't be cashed in for two or three years had been, he said, "a wonderful retention tool." When the bubble burst in 2000 and 2001, the awards given to people working 80-hour weeks were "suddenly worthless." Options are a bet on a future price. Their value depends entirely on where they start.
Two years later, in May 2003, Nicholas stepped down as chief executive — ostensibly, the reporting said, to work on his marriage. Alan "Lanny" Ross became CEO. The 2003 proxy letter to shareholders carries two signatures: Ross's, and "Henry Samueli, Ph.D., Chief Technical Officer and Co-Chairman of the Board of Directors." Scott McGregor took over as CEO in 2005 and would run the company for a decade. Samueli became chairman. The professor was, for the first time, the senior founder in the building. The loud half of the amplifier was gone.
Payback Time
Susan Samueli has said it simply: "The first thing that entered our mind when we went public was how to give some away." The foundation was set up in anticipation. "We set up the foundation right away knowing that we would immediately start giving back as soon as that stock had value," Henry said in 2019. "It was not something we had to think for a second about."
The first serious gift was not to a university. It was to a congregation in trailers.
When the family moved from Los Angeles to Orange County in 1995, they went looking for a Reform temple in the southern part of the county and drove past it — once, twice, three times — with the three girls in the car. Temple Beth El in Aliso Viejo was housed in a small cluster of portable buildings, easy to miss. "It was a shock to me that the only Reform temple was in trailers," Samueli said. "So in the back of my mind I said, 'Gee, we have to do something about that someday.' But obviously at that point I had no idea what would happen." In the spring of 1999 the Samuelis gave Temple Beth El $3 million toward a $15 million, 65,000-square-foot building a block away. Rabbi Allen Krause had expected his 550-member congregation to raise $2.5 million on its own. After the Samueli gift, the members matched it.
Then the requests came. "Overwhelmed," he said in April 1999 — "with requests for donations, board positions, you name it, they're just flowing in." Opera Pacific's executive director took him to his first opera in early March of that year; Samueli promptly gave $50,000 toward a production of Brundibar, an opera for children. The Orange County Performing Arts Center put him on its board explicitly to reach the region's new technology money; he had, the center's development chief said, "instant panache with high-tech companies that we are all courting." The Samueli Foundation would soon field more than twenty proposals a week.
On Tuesday, December 21, 1999, he announced the gifts that would put his name on two buildings. Thirty million dollars to UCLA's engineering school, the second-largest single cash gift in the university's history. Twenty million to UC Irvine's, the largest cash gift the campus had ever received. Both schools would be renamed for him. Both would fund Samueli scholarships for undergraduates, Samueli fellowships for graduate students, endowed Samueli chairs.
"It's payback time," he said. "UCLA has been very understanding about my starting a company. I hope to help promote the next guy who is going to start a Broadcom." And: "I won't let the money be wasted on bureaucracy or media buzz. I was once a struggling student and a young faculty member. I know what they go through."
The Times, to its credit, noted the other ledger. Over the previous five years, Broadcom had hired 80 percent of the electrical engineering graduates of UCLA's integrated circuits group. The two schools he was endowing were the two places where Broadcom competed hardest for talent. Philanthropy, the reporters wrote, "also represents smart business." Both things were true. The reverence and the wryness belong in the same paragraph, because the man himself seems to have held them there without discomfort. Gratitude is not less sincere for compounding.
The names multiplied: the Samueli Jewish Campus in Irvine, $25 million, in 2001. The Susan Samueli Center for Integrative Medicine at UC Irvine, from a $5.7 million gift in 2000. The 500-seat Samueli Theater at the Orange County Performing Arts Center, opened in 2006. A charter high school in Santa Ana for underserved and foster teenagers, the Samueli Academy, in 2013. In 2012, the couple signed the Giving Pledge. In 2017, $200 million to UC Irvine to create the Susan and Henry Samueli College of Health Sciences — at the time the largest gift in the university's history, and the occasion for a Los Angeles Times column by Michael Hiltzik headlined, bluntly, "A $200-million donation threatens to tar UC Irvine's medical school as a haven for quacks." The integrative-health mission, Susan's mission, had followed her from the nursery into a university's medical school, and the precision engineer had signed the check.
In 2019 came $100 million to UCLA, the largest gift the namesake school had ever received; by 2025 the couple's UCLA total was roughly $190 million and their UC Irvine total $307.7 million, the largest of any donor. In 2023, another $50 million to UCI's engineering school for three research institutes. "Engineers are hard-wired as problem solvers," he said. When IEEE gave him its Medal of Honor in 2025, with a newly enlarged $2 million prize, and when the Marconi Society gave him its prize in 2012, he turned the money around into an award for young innovators — the Marconi/Samueli Award for Innovation, inspired, the foundation said, by a seventh grader and a Graymark kit.
One name stands apart from all the others. In 2003, the Samueli Foundation gave Chapman University in Orange its Holocaust memorial library. It is not named for Henry, or for Susan. It is the Sala and Aron Samueli Holocaust Memorial Library. The engineering schools carry the son's name, a declaration of arrival. The library carries the parents' names, a record of what they arrived from. It is the only building in the catalogue where the giver is not the subject.
There is now, inevitably, a book — Henry Samueli: The Visionary Behind… — whose title gives the register of most writing about him: the visionary, the benefactor, the namesake. One UCLA parent who heard him address a graduating class wrote afterward, with the casual cruelty of the internet, that a student had confessed to having "no idea the guy was even alive." It is the fate of the very successful donor to become a facade — a name chiseled over an entrance, a person no one thinks of as still walking around. The facade is a kind of erasure, too.
What the Grant Dates Meant
A stock option is a right to buy a share at a fixed price at some future date. Its value depends on the gap between that fixed price — the exercise price — and wherever the stock is when you cash in. Set the exercise price on a day when the stock happened to be low, and the option is worth more the moment it exists. Set it on that day after the fact, and you have backdated it.
In March 2006, a report from the Center for Financial Research and Analysis listed Broadcom among seventeen companies "at risk" of having backdated option grants between 1997 and 2002. The company opened an internal review in May. In June, the SEC asked for information. On July 14, Broadcom estimated it would need to subtract $750 million from earnings; by September, it was at least $1.5 billion, "and could be substantially more." In December, the SEC opened a formal investigation. On January 24, 2007, Broadcom restated its results from 1998 through 2005 to include $2.24 billion in additional stock-based compensation expense. It was the largest restatement among the roughly 250 companies caught up in the backdating era.
The internal investigation, run by lawyers and forensic accountants hired by Broadcom's outside directors, laid most of the blame on Nicholas and on Ruehle, the former CFO. It concluded that Samueli had "reasonably relied on management and other professionals regarding the correct option accounting treatment and grant approval process." A corporate-governance firm, Proxy Governance, urged shareholders in April 2007 to withhold their votes for him anyway, calling the board's process "unsatisfactory and not sufficiently independent." On May 25, 2007, Samueli sat for his SEC deposition and said the sentence: "I was not involved in the actual granting process."
He had been. In January 2002, he would later admit, he twice helped decide the date on which options should be granted.
The machinery moved. In November 2007, Broadcom's former human resources director, Nancy Tullos, pleaded guilty and agreed to cooperate. In April 2008, Broadcom paid $12 million to settle an SEC suit against the company. In mid-April 2008, Nicholas checked into the Betty Ford Center for alcohol treatment. On May 14, 2008, the SEC sued Nicholas, Samueli, Ruehle and the company's general counsel, David Dull, alleging that they had systematically backdated 232 million options across as many as 88 grants over five years, June 1998 through May 2003. "A massive, five-year scheme," said the SEC's enforcement director, Linda Chatman Thomsen, "that involved fraudulent backdating of dozens of option grants, falsifying corporate records, intentionally false accounting, and lying to shareholders."
Nicholas and Samueli, the complaint noted, had formed a two-man committee that awarded 95 percent of the options. Neither had received any backdated options himself. Prosecutors' theory was that the founders had given in-the-money grants to employees and other executives — the Lamborghini owners in the Irvine parking lot — while avoiding the obligation to report the cost to shareholders.
Hours after the suit, Samueli resigned as chairman and took leave as chief technology officer. "I could not in good conscience allow today's unfortunate turn of events to become a distraction to the company I co-founded," he wrote. He would stay on as a technology adviser to the CEO. He would continue, he said, "pursuing energetically my family philanthropic endeavors as well as our other family business activities including the Anaheim Ducks." His lawyer, Gordon Greenberg, accused the SEC of "trying their case in the media."
On June 4, 2008, a federal grand jury indicted Nicholas and Ruehle. Nicholas was separately indicted on drug charges, with allegations that filled the tabloids of two continents. Samueli was named in the Nicholas-Ruehle indictment as an unindicted co-conspirator, identified as "H.S." On June 20, he was charged with a single count of making a false statement to the SEC. On June 23, three days later, he pleaded guilty in federal court in Santa Ana. The deal: five years' probation, a $250,000 fine — the statutory maximum — and a $12 million payment to the U.S. Treasury. In exchange, no securities-fraud prosecution. He was 53, and he became, the Times reported, the most prominent executive convicted in the federal backdating investigation.
Columbia law professor John Coffee offered the cynical forecast: "There are people who come back from criminal convictions, but usually only after a long period of penance and full admission of responsibility. But if he continues to give away a billion dollars in a thoughtful manner, I'm sure there are plenty of people out there who will forgive him." The NHL suspended him as an owner. Commissioner Gary Bettman's statement allowed that Samueli "has been an exemplary owner," and then imposed discipline anyway.
What happened next was not in anyone's forecast.
On September 8, 2008, Judge Carney refused the deal. A wealthy defendant writing a $12 million check to resolve his criminal exposure in an alleged $2.2 billion fraud, without cooperating against his co-defendants, looked to him like something worse than leniency.
— U.S. District Judge Cormac J. Carney, rejecting the Samueli plea agreement, September 8, 2008The court cannot accept a plea agreement that gives the impression that justice is for sale.
Samueli appealed. On September 24, 2009, the Ninth Circuit dismissed the appeal for lack of jurisdiction. He could now have withdrawn his guilty plea and gone to trial, or bargained again. He did neither. He let the plea stand, unsentenced, with the theoretical prospect of years in prison, and waited.
Which brings us back to the courtroom. Ruehle's trial; Nicholas's case severed; Samueli called by the defense under a grant of immunity; two days on the stand, Tuesday and Wednesday, December 8 and 9, 2009. He walked the jury through Broadcom's founding, the 16-hour days, the Morgan Stanley bankers who had recommended Ruehle as a man of "unshakable integrity," the retention tool that became worthless when the bubble burst. He admitted, again, the false statement. He explained, again, that he could not explain it.
On Wednesday, when he finished, Carney asked him to step down and face the bench. The statement to the SEC, the judge said, had been "ambiguous and evasive" — but it was not a material false statement, and so it was not a crime. "I have listened to your testimony and I can envision that this is actually a very truthful response," Carney said. "You are not going to be prosecuted for anything dealing with your S.E.C. testimony." The plea would be expunged. Double jeopardy, he believed, would bar the government from refiling.
Samueli bowed his head. "Thank you, your honor."
Then, to the judge: "That you truly understood what happened here means a lot to me. You have restored my faith in the criminal justice system."
Wayne Gross, a former chief of the U.S. attorney's office in Santa Ana, called it unprecedented: "I have never seen a judge dismiss a case against a defendant who had already pleaded guilty and who is not even on trial. Samueli was merely a witness in someone else's trial, which reflects the extent to which the judge was appalled." Outside, with Susan beside him, Samueli said, "I'm in shock." Susan said, "Henry's the most ethical man I ever met. The truth came out." Inside, the trial continued; one of the federal prosecutors, Andrew Stolper, would acknowledge that he had leaked information about the case to the press.
A week later, Carney threw out the backdating charges against Nicholas and Ruehle as well, citing a campaign of prosecutorial misconduct and witness intimidation. In his written findings on Samueli, he went further than any defense lawyer could have asked:
The uncontroverted evidence at trial established that Dr. Samueli was a brilliant engineer and a man of incredible integrity. There was no evidence at trial to suggest that Dr. Samueli did anything wrong, let alone criminal. Yet, the government embarked on a campaign of intimidation and other misconduct to embarrass him and bring him down.
The government's treatment of him, Carney wrote, was "shameful and contrary to American values of decency and justice." And then the sentence that closes the circuit eight years after a conference room with the shades drawn:
"Dr. Samueli now has a clean slate."
In January 2010, the drug charges against Nicholas were dropped. Broadcom ultimately paid $160 million to settle the related shareholder litigation. In December 2009, Samueli returned to the company as senior vice president and chief technical officer. The NHL had already reinstated him as an owner. He now appears in the National Registry of Exonerations — case number 11090 — alongside people who spent decades in prison for crimes they did not commit, a juxtaposition that is either a testament to the registry's rigor or a small absurdity, depending on where you stand.
It would be dishonest to pretend that the clean slate settles everything. The restatement was real; $2.24 billion of expense had not been reported when it should have been. The grants had been dated. The two-man committee had made 95 percent of them, and one of its two members was Samueli. What the judge found was that the professor had not lied criminally — that he had given an evasive, wrong, careless answer about a process whose accounting he had delegated to people he trusted. That is exoneration of a kind. It is also, for a man who chose his field because "it's exact," a strange place to have landed: acquitted because his statement was too ambiguous to be a lie.
The Hockey Owner Who Grows a Beard
The sporting life began, like so much in his story, with a management contract and a tenant. In 2003, the Samuelis bought the contract to run the Arrowhead Pond of Anaheim, forming Anaheim Arena Management LLC. The arena's largest tenant was the Mighty Ducks of Anaheim, a franchise the Walt Disney Company had created in 1993 to cross-promote a children's movie. In 2005, the Samuelis bought the team from Disney — for $75 million by one account, $70 million by the Orange County Business Journal's. In 2006, they dropped "Mighty" from the name and sold the arena's naming rights to Honda.
In 2007, the Anaheim Ducks won the Stanley Cup, the first California team ever to do it. That summer the Cup made the rounds of Newport-Mesa: a stop at Muldoon's, a stop at Fashion Island, a stop at the Orange County Performing Arts Center, where the 500-seat theater had opened the year before with the owners' name over the door. In 2008, the team was welcomed at the White House. In 2008, its owner was suspended by the league.
The Orange County Register, compiling a biographical box on him the week before the SEC suit, wrote that "despite his professorial demeanor, Samueli is a scrappy competitor." In 2015 the same paper noted that he was "known for maintaining a balanced rationalism. But he also is known for growing a playoff beard for luck." The man who credits Broadcom's existence to a window of timing he didn't choose, who says entrepreneurship "takes a lot of luck," lets his beard go during the playoffs. It is one of the few places in the record where the exact science admits it is not exact.
He has since sat on the executive committee of the NHL Board of Governors. In 2015 he bought the Norfolk Admirals of the American Hockey League and moved them to San Diego as the Gulls. That same year he announced a four-rink ice complex at Irvine's Great Park. The franchise that cost $70 or $75 million was valued by Forbes at $1.3 billion by 2025.
The team is the visible thing. The larger thing is the land around it.
The Company That Kept the Name
On May 28, 2015, Avago Technologies — a Singapore-headquartered chipmaker with roots in Hewlett-Packard's 1961 semiconductor division, spun out through Agilent and bought in 2005 by KKR and Silver Lake for $2.6 billion — announced it would acquire Broadcom Corporation for $37 billion, $17 billion in cash and $20 billion in stock. Avago was the nominal survivor. The combined company took the acquired company's name. Nicholas, on the phone with a reporter that day, was "like giddy": "I want to say this is the culmination of a life's work, but what we've really got now is a new point of departure."
The deal closed on February 1, 2016. Broadcom Corporation's shareholders held 32 percent of what became Broadcom Limited. Hock Tan, Avago's chief executive — an operator famous for buying franchises and running them with uncompromising discipline — became CEO. Samueli became chief technology officer and joined the board. The BRCM ticker was retired. The company kept AVGO.
What followed was not Samueli's company in any operational sense, and the people who know him are careful to say so. He described his role in 2015 as having changed: "I am not the one who invents the technology. We have tens of thousands of workers at the company, more than 800 of whom have Ph.D.s. They are the ones who do the inventing, and I serve as more of a mentor to them." In late 2018 he stepped back as CTO and became chairman of the board — a position UCLA's own board listing now records as running from December 2018 to June 2026. The Orange County Business Journal has described him, year after year, with the same phrase: "Remains confidante of CEO Hock Tan."
Under Tan, the company became something else. A $130 billion hostile bid for Qualcomm in November 2017 — Qualcomm, the old adversary with whom the original Broadcom had fought a four-year patent war that ended in April 2009 with Qualcomm paying $891 million — was blocked by presidential executive order on March 12, 2018, on national-security grounds, the company then being domiciled in Singapore. Broadcom reincorporated in Delaware and turned to software: CA Technologies for $18.9 billion in 2018, Symantec's enterprise security business for $10.7 billion in 2019, VMware for $69 billion, closed November 22, 2023. It drew antitrust orders from Brussels and a settlement with the FTC over contract terms that punished customers it deemed "disloyal." It also co-designed every generation of Google's Tensor Processing Unit and became one of the two or three indispensable suppliers of the artificial-intelligence buildout.
In December 2024, Broadcom became the twelfth company to pass a trillion dollars in market value. In April 2026, it passed two trillion. Fiscal 2025 revenue: $63.9 billion. Employees: 33,000.
Samueli's stake had been diluted to under 2 percent — 1.9 percent, by the 2025 proxy — and he had sold more than $5 billion of shares since 1998. It did not matter. Forbes had his net worth around $3.5 billion in 2020; by 2025 estimates ranged from $29 billion (the Business Journal) to $36.3 billion (Forbes). He displaced Donald Bren, the chairman of the Irvine Company, as the richest person in Orange County. In 2024, asked by the Business Journal whether he had ever imagined the company growing to its present size, he said, "Never in my wildest dreams."
He had once been the sole author of the fortune — the patents, the receiver architectures, the 95 percent. Now the fortune was mostly the work of a man he hadn't hired, buying businesses he hadn't founded, building chips for a technology he hadn't foreseen. He was rich the way a founder of a radio station might be rich if the station had been bought and turned into a television network and then a satellite company. The signal had gone out and been rebroadcast so many times that the original source was hard to hear. Of course, he had said it himself, about the radio, before any of it: he had built a thing that worked, and he did not know how.
Leave of Absence
In November 2024, Henry Samueli pledged 16.2 million shares of Broadcom common stock, worth roughly $2.7 billion at the time, to secure loans for a real estate project in Anaheim. He and Susan had been buying land around the arena since 2018 — parking lots, an office tower at 2400 East Katella for $72.9 million, a 12-acre industrial site near Angel Stadium for $64.5 million — well over $200 million of commercial property, assembled parcel by parcel. On it they are building OCVibe: a $4 billion, privately funded, 100-acre district with a 6,000-seat concert venue, 35 restaurants, 1,500 residential units (15 percent of them affordable), two hotels, a million square feet of offices, and twenty acres of public plazas and parks. Next door, they are spending another $1.1 billion renovating the arena itself — opera-box suites, a five-story south entrance, new escalators. They signed a 50-year lease renewal. There are no limited partners. There is no public money. They did not need to ask.
The purpose, as he explained it at an executive breakfast in May 2022, is retention. Young professionals leave Orange County for cities they find more exciting. The goal is "to create a magnet, so people stay here." The jobs exist; he wants "there's a great life behind those jobs."
It is the same idea as the stock options, and the same idea as the engineering gifts that funneled 80 percent of a UCLA lab into a Broadcom building. Make the place good enough that the people don't leave. A man whose parents crossed an ocean with nothing, who never left the Los Angeles basin for school, who never resigned the faculty post he took a pay cut to accept, has spent a fortune on the problem of why people go.
In June 2025, on the fiftieth anniversary of his bachelor's degree, he stood in Pauley Pavilion and gave the commencement address to the undergraduates of the school that bears his name. He told them not to catastrophize the future, that troubleshooting and problem-solving would survive whatever artificial intelligence did to their professions. "Don't view disruption as an exception, but rather view it as a rule," he said. "It will happen, so do your best to take advantage of it." It was good advice from a man whose own life had been disrupted on schedule — by the Holocaust before his birth, by a bubble, by a prosecutor, by a buyer from Singapore — and who had, each time, done his best to take advantage of it.
The honors have arrived in the order honors arrive for men in their seventies: the IEEE Medal of Honor, presented in Tokyo in April 2025; the National Inventors Hall of Fame in 2026, with Patent No. 5,754,591; the American Philosophical Society, founded by Benjamin Franklin, the same year. At the Hall of Fame announcement, he spoke about institutions. "A spirit of innovation rises from institutions such as UC Irvine, UCLA and the University of California system to the state of California and to our nation and the rest of the world."
The UCLA faculty page lists his research interests — "VLSI implementation of signal processing and digital communication systems, high-speed digital integrated circuits, digital filter design" — and his office, 58-121 in Engineering IV, and his phone, (310) 825-2647, as if a student might call during office hours about a problem set. The leave of absence is now three decades old. Nobody has asked him to come back, and he has never resigned.
And in a garage in Orange County — or there, at least, in December 1999, the month he gave away fifty million dollars and told two reporters where he kept it — sits a five-tube shortwave receiver, a Graymark 506B, ninth-grade work assembled by a seventh grader who soldered every night for a semester, carried it to class, watched his teacher plug it into the wall, and heard music come out of the air.
Part IIThe Playbook
What follows is an attempt to extract usable principles from a career that its subject mostly attributes to timing and luck. That attribution is itself the first lesson: Samueli is unusually honest about what was outside his control, which made him unusually disciplined about what was inside it. The principles below are grounded in the evidence of Part I. Some are about building technology companies; several are about the less glamorous work of being the second founder, the quiet one, the one who stays.
The principles
- 1.Choose the field that matches your temperament.
- 2.Ask for the harder kit.
- 3.Keep the leave of absence open.
- 4.Let the lab be the farm team.
- 5.Fund the startup with someone else's mission.
- 6.Sell the spec before the silicon.
- 7.Be the positive side of the push-pull pair.
- 8.Name the bad quarter out loud.
- 9.Never delegate the thing you'll be asked about under oath.
- 10.Narrow the giving, then go hunting.
- 11.Stay as confidant, not as operator.
- 12.Respect timing; don't worship it.
H
The Samueli Arc
Key decisions and inflection points, 1966–2026.
1960s
Seventh grader at Bancroft Junior High builds a Graymark 506B shortwave radio instead of the assigned crystal set.
1980
Completes Ph.D. at UCLA under Alan Willson; joins TRW's military broadband group.
1985
Dean Frank Wazzan breaks the no-alumni rule to hire him full-time; Henry Nicholas becomes his first Ph.D. student.
1988
Begins consulting for PairGain; its first DSL transceiver (1991) takes 80% of its market.
1991
Broadcom founded in August with $5,000 from each founder.
1995
Moves company to Irvine; takes indefinite leave from UCLA.
1998
Broadcom IPO in April; Samueli Foundation established.
1999
$30M to UCLA and $20M to UC Irvine; both engineering schools named for him.
2005
Buys the Mighty Ducks from Disney.
2007
Ducks win the Stanley Cup; Broadcom restates $2.24B in options expense.
2008
SEC suit; resigns as chairman; pleads guilty to one false-statement count; Judge Carney rejects the deal.
2009
Carney dismisses the case from the bench; Samueli returns as CTO.
2016
Avago completes $37B acquisition; Samueli becomes CTO of the combined company.
2018
Steps back as CTO; becomes chairman.
2025
IEEE Medal of Honor; self-funded $4B OCVibe and $1.1B arena renovation underway.
Principle 1
Choose the field that matches your temperament
Samueli did not fall in love with electrical engineering in the abstract. He fell in love with one sub-discipline, digital signal processing, for a reason that is entirely about his own psychology: "I'm a perfectionist-type of person… the beauty about digital signal processing is it's exact." Most career advice treats temperament as something to overcome. He treated it as a selection criterion and picked the corner of the field where his compulsion for exactness was an asset rather than a drag.
The payoff compounded. DSP was precisely the discipline that would make cheap, high-speed digital receivers possible once CMOS got good enough, and a perfectionist who had spent a decade inside it was positioned to lead when the window opened. His dissertation on overflow oscillations — the obscure question of what happens when a digital filter exceeds its numerical limits — is the kind of thing only a temperamentally matched researcher spends years on, and exactly the kind of deep, unglamorous knowledge that makes a chip work in the field.
Tactic: Write down the specific quality of work that gives you pleasure — exactness, speed, persuasion, ambiguity — and choose the narrowest niche of your field where that quality is the bottleneck.
Principle 2
Ask for the harder kit
The defining anecdote of his childhood is not that he built a radio; it is that he refused the assigned radio. The crystal set was boring. The shortwave set was a ninth-grade project, and the teacher said no. Samueli "persisted and eventually prevailed," then spent a full semester soldering hundreds of joints every night. The teacher's verdict — "nobody gets it right the first time" — mattered less than the habit it reinforced: ask for the harder project, then carry the full cost of having asked.
That habit reappears in the company's go-to-market. Broadcom routinely told the largest customers in a market that it could build "something they thought was impossible to build." It also reappears, darkly, in the plea: he let a guilty plea stand rather than withdraw it after the deal collapsed, accepting more risk instead of less. Asking for the harder kit is a disposition, and dispositions don't switch off when the stakes change.
Tactic: When handed a scoped assignment, propose one materially harder version and commit in writing to finishing it on the original deadline.
Principle 3
Keep the leave of absence open
Samueli straddled for an unusually long time. He taught part-time while at TRW. He consulted for PairGain while a professor. He co-founded Broadcom in 1991 and did not go full time until 1995, four years in. When he finally did, he took a leave, not a resignation — and he has never converted it. Thirty years later, he is still on the UCLA roster.
This was not indecision. It was optionality managed as an engineering margin. The academic post funded his life and gave him a pipeline of graduate students to hire as part-timers; the company did not need to pay a professor's salary until it could afford one. When Wazzan granted the indefinite leave, he gave Samueli the one thing most founders lack: a respectable path back. That safety let him take the startup risk without the desperation that warps decisions.
⇄
Straddle vs. Leap
How Samueli's sequencing differed from the standard founder script.
| Standard script | Samueli's approach | Phase |
|---|---|---|
| Quit the day job to signal commitment. | Kept the professorship four years after founding. | Scan |
| Raise venture capital to hire fast. | Funded R&D with defense contracts; never took VC. | Try |
| Recruit through networks and headhunters. | Hired his own graduate students part-time, then full-time. | Try |
| Resign once the company works. | Took an indefinite leave, never closed. | Scale |
| Exit when acquired. | Stayed as CTO, then chairman, of the acquirer. | Scale |
Tactic: Before you leave an institution, negotiate a formal leave rather than a resignation; the cost to them is low and the value to your risk tolerance is high.
Principle 4
Let the lab be the farm team
Over the five years before 1999, Broadcom hired 80 percent of the electrical engineering graduates of UCLA's integrated-circuits group. The professor was, in effect, running a talent pipeline from his own lab into his own company. Then he gave $50 million to the two engineering schools where Broadcom competed hardest for graduates, and the Los Angeles Times correctly noted that it was smart business as well as gratitude.
The lesson is not cynicism. Samueli's sentiment was real — "I was once a struggling student and a young faculty member" — and the gifts funded scholarships, fellowships, and chairs that benefited students who never worked at Broadcom. But the alignment was intentional. He built a regional talent ecosystem in Southern California at a time when, as one Irvine executive put it, "everyone else was running straight to Stanford." The philanthropy and the recruiting were the same investment, pointed at the same people.
Tactic: Identify the two or three institutions that produce the talent you most need and invest in them before you need to, in ways that would be worth doing even if you never hired from them.
Principle 5
Fund the startup with someone else's mission
Broadcom's first revenue came from small military contracts, work both founders knew from TRW. "These projects funded our R&D, and we gained more and more knowledge," Samueli said, and the military work produced "a good two-year lead over the competition." The company never spent beyond the terms of its contracts, never took venture capital, and was profitable from inception.
The strategic insight is that a customer with a demanding, well-funded mission adjacent to yours can pay for your learning curve. Defense communications needed exactly the high-speed digital modem technology that consumer broadband would later need, years before the consumer market existed. By the time cable operators wanted chips, Broadcom had already solved the hard problems on someone else's dime — and kept its equity and its independence, which, as Forbes noted, "left Nicholas and Samueli free to run the company in their own unorthodox way."
Tactic: Find a customer whose current need is a demanding version of your future market's need, and price contracts to cover your R&D rather than to maximize margin.
Principle 6
Sell the spec before the silicon
Broadcom's playbook, as described in 2000: pick the largest customer in each networking-chip market, design the best chip for that customer, build it cheaper than the incumbent — and win order commitments before building a prototype. "We show them the specs and convince them that we can build something they thought was impossible to build," Nicholas said.
This only works with deep technical credibility, which is why the founders' Ph.D.s and the company's hiring preference for academics were commercial assets, not ornaments. Samueli himself named credibility as the start-up's central problem: "The biggest challenge that a start-up has is credibility that can really deliver the product that's reliable and can be manufactured." A commitment from the largest customer converts credibility into revenue visibility, de-risks the design spend, and gives the startup a reference account that smaller customers will follow.
Tactic: Before building, take a detailed specification to the single largest potential buyer and ask for a conditional purchase commitment; if you can't get one, your spec isn't credible enough yet.
Principle 7
Be the positive side of the push-pull pair
Nicholas's own description of the partnership — he could push engineers to tears only because Samueli was "on the positive side" — is a rare candid account of how founding duos actually function. The Times described Samueli as "a calming father figure to his beloved engineers" who "rebuilt their confidence and set them back on task." Without Samueli, Nicholas's intensity would have burned the organization out. Without Nicholas, Samueli's deliberateness might have left the company too careful to take on Lucent and Intel.
2
The Two Henrys
Complementary roles in Broadcom's founding partnership, as described in contemporary reporting.
| Dimension | Henry Nicholas | Henry Samueli |
|---|---|---|
| Role | CEO; ran the business, drove to win | CTO; inspired the engineers |
| Described as | "The fire," "brash," "hyper-energy" | "Deliberate," "reserved," "taciturn" |
| First purchases after IPO | Ferrari, Lamborghini, Harley, jets | House, kids' toys, a Porsche for an employee; kept leased Cadillac |
| Effect on engineers | Pushed them hard enough to make some cry | Rebuilt their confidence |
| Stated creed | "I always want to win." | "The billionaire thing doesn't register." |
The more useful observation is about role choice. Samueli could have competed with Nicholas for the spotlight. He didn't; he was, in public, "a nearly invisible counterpoint." That invisibility was a deliberate contribution to the system, and when Nicholas left in 2003, the counterpoint became the steadying center.
Tactic: In any leadership pair, explicitly decide who carries pressure and who carries repair, and resist the temptation for both of you to play the same role.
Principle 8
Name the bad quarter out loud
In June 2001, with the stock down from nearly $275 to around $30, Samueli did not soften the news to his top engineers. "You can do back flips, and you're still not going to salvage" the quarter. Layoffs were coming, and he wanted them done quickly: "We have to do it and get it out of the way and start with a clean slate. We can't keep talking about negative things."
Two moves are packed into that meeting. First, an unambiguous verdict that ends the bargaining — the engineers had been proposing salary sacrifices and equipment sharing, and he closed that loop. Second, a pivot to morale as the explicit next task. Downturns kill companies less often through the cuts themselves than through the prolonged uncertainty around them. His instinct, the opposite of his deposition answer two decades later, was clarity.
Tactic: When a bad outcome is certain, state it plainly to the people affected, set a short deadline for the hard action, and announce the recovery goal in the same meeting.
Principle 9
Never delegate the thing you'll be asked about under oath
The backdating saga is the costliest episode of Samueli's life, and its lesson is narrow and practical. He sat on the two-man committee that awarded 95 percent of the options. He relied — "reasonably," the internal investigation concluded — on management and professionals for the accounting and the approval process. He had no formal accounting training. And when asked about his involvement, he gave an answer that was, in the judge's word, "ambiguous and evasive," because he had mentally filed the process under someone else's jurisdiction.
The exoneration was complete; the cost was not. He lost his chairmanship for eighteen months, was suspended by the NHL, carried a felony plea for a year and a half, and watched the company record the largest restatement of the era. The principle is not "be suspicious of your CFO." It is that a founder who holds signing authority over something owns the understanding of it, regardless of whose committee it formally belongs to. The perfectionist's blind spot was the domain he didn't consider his.
Tactic: List every decision for which your signature or vote is required, and for each one, be able to explain in two sentences how the downstream accounting and approval actually work.
Principle 10
Narrow the giving, then go hunting
Samueli admits the early philanthropy was messy. Requests were "flowing in"; the foundation received more than twenty a week; he joked that he had never met a nonprofit unworthy of support. "How do you learn to say 'no' properly? Narrowing down your mission is one way," he said in 2019. "Then put together a plan for how we're going to do it and proactively go out and seek the nonprofits you want to support. That way, you're in control of the giving."
The mature portfolio reflects that discipline: STEM education (the engineering schools, the Broadcom MASTERS middle-school competition, the Samueli Academy), integrative health (Susan's field), youth services, Jewish culture. Each maps to a biographical fact — the radio, the nursery, the parents' rescue by aid organizations. The narrowing did not reduce the scale; it increased it, because concentrated giving builds institutions rather than scattering grants. Even the middle-school focus has a reasoned basis: the foundation targets the age at which STEM interest drops off, which is also the age at which Samueli asked for the harder kit.
Tactic: Define your giving around three or four causes tied to your own biography, publish them, and spend the majority of your effort originating gifts rather than responding to requests.
Principle 11
Stay as confidant, not as operator
After the 2016 acquisition, Samueli did something many founders cannot: he stayed inside a company he no longer ran and did not try to run it. He described himself as a mentor to the inventors rather than the inventor. He stepped back as CTO in 2018 and became chairman. The local business press uses the same phrase every year: he "remains confidante of CEO Hock Tan."
The financial result was extraordinary. His stake fell below 2 percent while his wealth grew roughly tenfold, because Tan's strategy — software acquisitions, custom AI silicon — multiplied the value of what remained. A founder who insisted on preserving the original Broadcom's identity might have fought the transformation. Samueli kept the name on the door, gave counsel, and let the operator operate.
Tactic: If you sell to a stronger operator, define your continuing role as advisory in writing on day one, and measure your success by the acquirer's results rather than by how much of your original company survives intact.
Principle 12
Respect timing; don't worship it
"Timing is everything," Samueli tells students. Ten years early, Broadcom fails; five years late, it fails. "To be successful, it takes a lot of luck." It is unusual humility from a founder, and it is also a strategy. Because he believed timing was decisive, he positioned for it years in advance — a DSP doctorate, CMOS research, defense contracts that built the capabilities consumer broadband would need — so that when the window opened, Broadcom was already standing in it.
The same logic governs his late-career bets. He told students in 2014 that Moore's Law was slowing and that the next twenty years belonged to sensors integrated with communications, especially in medicine. He told the class of 2025 to treat disruption "as a rule" rather than an exception. You cannot choose the moment. You can choose to be prepared, financially independent, and technically ahead when it arrives — and, as he did with a playoff beard, you can admit what remains outside your control.
Tactic: For the market you want to win, identify the enabling condition that hasn't arrived yet, and spend the waiting years building the capability that condition will make valuable.
For readers interested in a fuller biographical treatment, Henry Samueli: The Visionary Behind… collects the arc from Buffalo to Broadcom.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
— Henry Samueli, announcing $50 million in gifts to UCLA and UC Irvine, December 1999It's payback time. UCLA has been very understanding about my starting a company. I hope to help promote the next guy who is going to start a Broadcom.
— Henry Samueli, National Inventors Hall of FameWe didn't push the technology out there, it was pulled out of us because the demand was so great.
— Henry Samueli, testifying in the trial of William Ruehle, December 2009I made that false statement and I accept it, but I bang myself in the head for making such a stupid statement.
— Henry Samueli, to PCMagI really do believe that engineers should run the world. I wish all politicians had an engineering background.
— Henry Samueli, UCLA Samueli undergraduate commencement, June 2025Don't view disruption as an exception, but rather view it as a rule. It will happen, so do your best to take advantage of it and be at the forefront.
Maxims
- Pick your field by temperament. Samueli chose digital signal processing because "it's exact," and the match between disposition and discipline became his edge.
- Refuse the easy kit. The crystal radio was the assignment; the shortwave set was the education.
- Leave, don't resign. An open door back to a respected institution buys the calm needed to take real risk.
- Let customers fund your learning curve. Defense contracts paid for the R&D that consumer broadband would later reward.
- Sell the impossible on paper first. A credible spec and a committed anchor customer are worth more than a prototype nobody ordered.
- Choose your half of the partnership. One founder pushes, one founder repairs; a company needs both and rarely survives two of the same.
- End the bargaining quickly. In a downturn, a clear verdict and a short deadline do less damage than hope.
- Own what you sign. Delegating the mechanics never delegates the responsibility to understand them.
- Give in a narrow beam. Concentrated, proactive philanthropy builds institutions; reactive giving scatters them.
- Prepare for the window you can't choose. Timing is luck; being ready when it opens is not.
