A lightbulb ought to have modest ambitions. Illuminate a room. Go dark when asked. Thomas (TJ) Jermoluk and Jim Clark encountered one with an additional requirement: a password. They were developing technology for home automation, and the business of switching on a light had acquired the habits of a website. Before there could be illumination, there had to be authentication.
Jermoluk has described that obstacle as the beginning of the idea behind Beyond Identity. The scale of the eventual company makes the starting point rather pleasing. After graphics workstations, cable networks and years in venture capital, here was a familiar computer problem attached to a household object. A lamp had become an argument.
The irritation was specific. Passwords had travelled into another corner of life, carrying their old complications with them. Jermoluk and Clark began asking whether the cryptography already used to protect internet communications could reach the individual user and device. The question became a business. The lightbulb, one hopes, eventually became a lightbulb again.
Computer time after dark
Long before that project, Jermoluk’s access to computing came with a timetable. Growing up in Hawaii, he learned programming through an 11th-grade math teacher who arranged rented nighttime access to a publicly available computer. At Virginia Tech, he joined the night-owl computing culture in McBryde Hall. He graduated in computer science in 1978 and completed a master’s degree in 1979.
There is a useful distance between those two experiences: booking time on a scarce machine and later trying to remove an unnecessary step from an ordinary person’s device. Computing became more available. The work of making it usable continued.
Jermoluk credits his school, ‘Iolani, with giving him a foundation for further study and a work ethic that carried into business. That account is quieter than the usual story of a founder discovering destiny. Teachers gave him something to work with. He kept working.
He went into research at Bell Laboratories and then development at Hewlett-Packard. His background spanned hardware and software, a practical education in the places where one system has to meet another. It helps explain why his career can cross markets without seeming to change its underlying subject: the engineering that lets people do something they previously could not.
The pictures, then the pipes
Jermoluk joined Silicon Graphics in 1986 as an engineer. By 1994, he was its president and chief operating officer. The company made visual computing systems; his responsibilities grew to include the organisation that designed and sold them. Moving into management meant being answerable for what happened after an architectural drawing left the whiteboard.
One of his recollections captures the attraction of the place. An uncomfortable executive interview gave way to a technical conversation with Clark that lasted four hours and filled eight whiteboards. A working relationship began in the pleasure of having enough to talk about.
Their subsequent partnership had room for different responsibilities. Clark supplied entrepreneurial pressure; Jermoluk took on the operating details. Hiring, finances and execution belong to that second category. These are the tasks that determine whether an idea can survive the less glamorous hours of a working week.
In the summer of 1996, Jermoluk left SGI for @Home. The new job concerned a different constraint: getting internet access through the capacity of cable television networks. A graphics machine might make a screen more useful. A faster connection could make the network beyond that screen more useful too.
“It’s an opportunity to build another thing,” he said at the time. The sentence has little ornament. It describes a motive that does not require the next company to resemble the last one. There was another system to put together.
@Home’s proposition required relationships with cable companies as well as engineering. Distribution was part of the product’s practical existence. Having the network was one problem; arranging for households to reach it was another. The operator’s job sat at the junction.
The broadband chapter also resists a tidy success story. In January 1999, @Home announced its acquisition of the Excite internet portal in a stock deal valued at roughly $7 billion. The ambition expanded from carrying internet traffic to supplying the destination people would see when they arrived.
The combined company struggled. Portal traffic and advertising did not make the financial picture behave as hoped, and relationships among the cable partners became difficult. George Bell succeeded Jermoluk as CEO in January 2000. Jermoluk left the chairman role that May. Excite@Home filed for bankruptcy in September 2001.
That ending belongs in his story. Cable broadband endured; this particular business did not. It is possible to help advance a useful technology and participate in a corporate strategy that ends badly. The distinction is part of the history, rather than an inconvenient footnote to it.
A secret is a strange kind of key
Venture capital followed. Jermoluk became a general partner at Kleiner Perkins, and his later career included leading Hyperion Development Group and founding other technology businesses. The work changed his position in the room. Instead of presenting only his own company’s plans, he could evaluate and support someone else’s.
“I’m a little-company guy; it’s a more comfortable environment for me,” he said in 2001. That preference makes sense beside the demands of @Home’s cable relationships. A small company can still have difficult decisions. It generally has fewer institutions that must be persuaded before a decision becomes action.
Beyond Identity brought him back to the founder’s side of the table. Established in 2019, it publicly launched in April 2020 with $30 million in Series A funding from Koch Disruptive Technologies and New Enterprise Associates. Jermoluk was co-founder and CEO; Clark was co-founder and chairman.
The product’s starting point was the shared secret. In an ordinary password arrangement, a user knows something that a service can check. Someone who obtains that credential can attempt to present the same proof. The system has to distinguish the person from a usable copy of the person’s secret.
Beyond Identity used public-key cryptography and certificates to bind authentication to a device. The private key remains with that device; the service can verify the proof without asking the user to transmit a reusable password. Device-security checks add another question: whether the machine requesting access meets the organisation’s requirements.
Jermoluk discussed the approach alongside cryptographers Martin Hellman and Taher Elgamal in 2020. His description of the company as standing “on the shoulders of giants” was an acknowledgement of technical ancestry. The business depended on taking established cryptographic ideas somewhere they could become more useful.
That is a different kind of invention from announcing an entirely new principle. An existing mechanism can become a new product when devices, infrastructure and deployment methods finally allow it to travel further. There is plenty of engineering in that journey.
He also pushed for care with the word passwordless. In a 2021 essay, he distinguished removing passwords from merely hiding them behind a more convenient interface. His argument extended to account recovery and the application database. A cleaner-looking login could leave the underlying credential problem intact.
The concern is revealing. The surface experience matters, but an engineer wants to know what remains underneath it. A customer might see one fewer box to fill in. Jermoluk wants the design to remove the secret that made the box necessary.
The money has a job to do
In February 2022, Beyond Identity announced a $100 million Series C led by Evolution Equity Partners. Total funding reached $205 million, and the announced valuation was $1.1 billion. These were figures for the company at that moment, with money intended for research, development and international expansion.
Jermoluk’s own response to the round was brisk, with a pointed observation about why investors had supplied the money. It was a useful thought to place near a valuation. Investors were financing a proposition about what authentication could become. The announcement did not finish the work.
“Nobody gives you $100 million because they think you're cool.”TJ Jermoluk, February 2022
His argument for the money emphasised preventing credential attacks rather than relying entirely on detecting and responding to them. The priorities were architectural: change the conditions under which someone can enter. It is the sort of question an operator can attach to product development, hiring and a budget.
There is a consistent practical tension here. A technology must be secure enough to justify using it and manageable enough for an organisation to adopt it. The successful demonstration is followed by the deployment, the integration and the person who has to administer the system on Monday.
Back to the machinery
Today, Beyond Identity lists Jermoluk as a co-founder on its board. Jasson Casey is CEO, and Clark is chairman. The operating title has changed; the company remains part of the career.
Another current connection leads back toward hardware. In September 2026, AI inference company Positron announced an $875 million financing at a $5 billion valuation and named Jermoluk among the directors joining its board. Positron’s board page identifies him with the Jim Clark Family Office. Clark participated in the financing.
Inference means running an AI model to produce an answer. For Positron, the practical problem is the cost and efficiency of the machines doing that work. Jermoluk’s place on the board brings the story back to computing infrastructure, where his earlier engineering career began.
Seen across those decades, the sequence has a certain order. Make the picture possible. Improve the connection. Reconsider the proof required to get inside. Then attend to the hardware serving another generation of applications.
The household light was a small object carrying a large assumption: that a password was simply the price of getting a machine to obey. Jermoluk’s next company began by examining that assumption. After all the networks, titles and financing rounds, the request was still wonderfully ordinary. Please turn on the light.