A famous customer is a persuasive thing. Put its name on a slide and a young company seems a little older, a little safer, a little more likely to survive the next question. At GraphDive, Shahram Seyedin-Noor once accepted that bargain: his startup did a project for a prominent Fortune 500 business for free. The attraction was the name. The problem, as he later acknowledged, was that the customer would have paid.
There is a small comedy here. A company devoted to extracting commercial intelligence from data had briefly misread the commercial intelligence sitting across the table. The customer had a budget. The startup had useful work to sell. Somehow, admiration had intervened between the two.
Seyedin-Noor's conclusion was admirably billable: “If your product has value, you need to charge for it.” The admission makes a better introduction than a string of job titles. He has accumulated those too: attorney, investment banker, finance executive, co-founder, chief executive, investor. The interesting part is what happened when the credentials met an actual customer.
A lawyer arrives during the party
His route into Silicon Valley began far from its conference rooms. Born in Iran, Seyedin-Noor came to the United States as a child. In a 2025 video interview, he described his upbringing as a refugee. The public arc that follows runs through an economics degree at Pomona College, graduation in 1996, and a JD from Harvard Law School in 1999.
At Harvard, he was editor-in-chief of the Harvard International Law Journal, leading a membership of 175. His LinkedIn profile also lists a published legal article about the prosecution of Augusto Pinochet. Before he spent his working life deciding which new businesses deserved capital, he had spent time with questions of jurisdiction, accountability and the machinery of institutions.
In 1999, he joined Wilson Sonsini. His identical twin brother started at the firm with him, an unusually literal version of bringing a familiar face to work. Shahram entered Alan Austin's group during the closing stretch of the dot-com boom. He remembers working late on IPOs over the winter holidays while the Nasdaq kept rising.
With an economics degree behind him, he thought the market looked like a bubble. A partner's response supplied the office nickname that the moment deserved: “Ha! Go back to your cave, you bear!” He recalls the exchange with affection. It is an appealing early scene: a junior lawyer arriving at the party and wondering about the bill.
“Ha! Go back to your cave, you bear!”
A partner's reply to his dot-com bubble warning, recalled in 2023
Several ways to read a business
The next stages were less linear than the familiar climb up a single firm's letterhead. His career included Goldman Sachs, Cooley and Bank of America. He advised technology businesses in investment banking and worked as a corporate lawyer before moving into operating roles. At NextBio, he became founding CFO and vice president of corporate development.
Those jobs offer different views of the same company. An attorney reads the obligations. A banker considers a transaction. A finance executive has to keep the enterprise funded while the work continues. For a future founder, there is practical value in having occupied more than one of those chairs.
He later described governance, negotiation and financial planning as business fundamentals that receive too little attention. It is an unfashionably useful list. A board meeting rarely makes a thrilling product demonstration. A badly managed board can nevertheless determine what happens to the product.
By the time he started GraphDive with Sina Sohangir in 2011, Seyedin-Noor had experience inside an analytics company as well as around financing tables. Sohangir, the co-founder and CTO, brought a Stanford PhD. Together they pursued a commercial question created by the spread of social login: what could a business learn from the information its visitors chose to share?
The customer behind the click
GraphDive was based in Menlo Park. Its Social API used social information to provide businesses with ranked interests, inferred demographics and personalized recommendations. The intended buyers included businesses in retail, travel and media. A site visitor could arrive as a relatively anonymous session and, after social sign-on, become someone a merchant could serve more specifically.
That was the commercial premise. More information should make a recommendation less arbitrary. Instead of treating every arrival as interchangeable, a business could sort audiences and tailor what it offered. For Seyedin-Noor, the test of the analysis was whether it helped customers sell.
In October 2012, GraphDive disclosed more than $1 million in funding, with Crosslink Capital and Correlation Ventures among its backers. The money was intended for product development and expansion as customer use grew. In September 2013, it raised a further $2 million in a round led by Start Capital, with participating angels including Pejman Nozad, Ullas Naik and Naguib Sawiris.
Walmart and Lyft were among GraphDive's reported customers. The names mattered, but a list of customers and a working business are different measurements. The former can help open a meeting. The latter depends on repeated use, reliable delivery and someone being willing to pay again.
The architecture also made a distinction worth retaining. Social login supplied an input. Analysis turned it into an estimate. A business then chose an action. Collapsing those stages into a single promise of knowing the customer would have made the pitch tidier and the product harder to judge.
Three lessons without a victory lap
Asked about mistakes at GraphDive, Seyedin-Noor named premature feature releases, paid platform memberships and the unpaid project for the large customer. The list is revealing for its lack of exotic villains. These were decisions a management team could revisit.
His complaint about early releases was specific to enterprise software: features needed testing in actual beta deployments. A customer fitting software into an existing operation has its own deadlines and dependencies. The founder's satisfaction at shipping something does not settle whether the buyer can rely on it.
He was also skeptical of paying to belong to another company's platform ecosystem. Membership is easy to count; productive distribution takes more examination. His objection invites a question any small business can ask of a partnership: what changes after the fee has been paid?
Then there was the free work. Large companies can offer a young supplier prestige without making prestige spendable. In his account, the customer eventually paid. The correction was available. What makes the episode useful is his willingness to identify the original concession as unnecessary.
A million-dollar starting point
GraphDive was acquired by Sociable Labs in 2014. In 2017, Seyedin-Noor founded Civilization Ventures. The firm's own account describes its beginning as a $1 million pilot seed fund. Its current website reports more than $200 million in capital under management and partnerships with over 70 companies. These are firm-level figures, rather than measures of his personal wealth.
The idea he calls “mentor capital” is to help founders develop into the responsibilities of running their companies. Civilization describes its approach as supporting founders rather than replacing them. The premise gives operating experience a second use: mistakes made in one company can become timely questions in another.
A founder may understand an invention long before learning how to negotiate a financing or organize a board. An investor who has handled those problems can bring something concrete to the conversation. The usefulness lies in the particular problem solved, which is harder to advertise than a promise of being supportive.
Even the administration of his own firm became a question of where time went. He switched from a traditional accounting provider to Carta for fund administration after finding the earlier arrangement's invoices unpredictable and its incentives poorly aligned with his firm's needs. Accounting, investor communications and reporting still had to happen. He wanted more time for working with founders.
That choice has a pleasing continuity with his earlier work in finance. Someone must mind the back office, especially when everyone would prefer to talk about the future. Choosing how that work gets done is part of running a venture firm, with consequences for the hours available to everyone else.
Experience that earns its keep
In a 2024 conversation about building and backing startups, Seyedin-Noor argued for taking responsibility even when an investor holds a small stake or lacks a board seat. Ownership of a problem, in that account, does not have to wait for a more impressive title. There is still work available to do.
That is a useful way to read his changes of occupation. The legal training remains relevant when agreements are negotiated. The finance work matters when resources have to last. The GraphDive chapter supplies the memory of being the person trying to turn useful technology into a paying relationship.
His current title is founder and managing partner of Civilization Ventures. GraphDive belongs to an earlier chapter, but it preserves something that a current biography can smooth away: the founder learning in public, with a recognizable customer and a preventable mistake.
A large logo had looked like payment. Experience taught him to send an invoice.
Follow the conversation
Interviews, professional profiles and further reading.