Mar 8, 2024,06:30am EST
In the summer of 2022, Garry Tan embarked on a journey from San Francisco to the Cotswolds, a serene region of England located about two hours west of London, for a pivotal meeting with his mentor, Y Combinator cofounder Paul Graham.
Tan, a seasoned participant in the YC program who transitioned from founder to partner before venturing out to establish his own venture capital firm alongside Reddit cofounder Alexis Ohanian, had ascended to the upper echelons of Silicon Valley’s power structure. He held the reins at Initialized Capital after Ohanian’s departure, secured a spot on the Forbes Midas List of top tech investors, and cultivated a substantial following on his YouTube channel.
During their rendezvous seven years later, Graham, an influential figure from the dotcom era who had evolved into a startup sage, proposed a new mission to Tan: to reinvigorate Y Combinator to its former grandeur. At 59 years old, Graham conveyed to the 42-year-old Tan that assuming the role of president and CEO of Y Combinator could expedite the arrival of a promising future.
The opportunity presented was nothing short of extraordinary. Y Combinator, renowned for nurturing startups like Airbnb (valued at $106 billion), Stripe ($65 billion), and DoorDash ($53 billion), sought a leader as Geoff Ralston, a longstanding associate of Graham, prepared to step down. Tan perceived it as a once-in-a-lifetime chance to guide the “fountain of prosperity for innovation and venture,” as he expressed to Forbes upon announcing his new position.
However, YC’s illustrious reputation had sustained a few blows after years of virtual batches, with a record number of over 400 companies undergoing the program concurrently. With a 7% equity stake for $125,000, plus an additional $375,000 convertible into future funding, Y Combinator’s participation didn’t come cheap, especially as its educational resources were freely available online. While some competitors like Techstars retreated, a fresh wave, from the HF0 hacker house to Sequoia’s Arc, emerged to vie for talent. Moreover, fledgling founders in burgeoning sectors like generative AI opted to forgo accelerators altogether, choosing to secure substantial funding from day one.
Amidst the slightest hint of wavering, detrimental to YC’s aspiration to remain relevant for generations, Tan emerged as the ideal catalyst—a disruptor from within the fold. By January 2023, he became arguably the most prominent VC firm leader to step away while at the pinnacle of his career.
Tan embarked on reshaping YC into a “return to roots,” revamping its batches from admissions to social activities to reclaim its status as a hub for early-stage, technical founders eager to grasp the fundamentals of company-building over three months. Operating from a new headquarters in San Francisco’s Dogpatch neighborhood, Tan projected YC’s influence within the city more boldly than ever before, alongside his personal endeavors, through increased public content, events, and local political activism.
By abruptly dismantling its Continuity fund—an eight-year-old growth-stage focused group within YC with significant investments in its brightest alumni companies—Tan executed an unconventional and resolute move. While celebrated by some as a prudent strategy, it stunned others with its severity, leaving certain founders—typically YC’s most prized assets and advocates—disgruntled.
However, Tan’s ambitions extend beyond Y Combinator. Simultaneously, he strives to revive San Francisco, his hometown, to the prosperity he believes it has lost. As Tan works towards streamlining Y Combinator’s programs and elevating its public profile, he spearheads a movement to reform the city’s government under moderate Democrat leadership, earning accolades from some and critiques from others.
While opinions about Tan’s approach are varied, his tenure thus far has been marked by vigorous debate and transformation. Founders, investors, and peers alike share strong viewpoints on his leadership, with some cautioning against a politicized and confrontational trajectory for YC, while others commend his intellectual prowess and founder-centric approach.
Ultimately, Tan’s tenure at Y Combinator is part of a constant evolution and experimentation within the institution, supported by figures like Altman, whose initiatives Tan is now revising. As one batch founder aptly noted, “YC feels very much back.”
Y Combinator cofounders Paul Graham and Jessica Livingston held court for its entire founder batch in February, answering questions on artificial general intelligence and how college dropout founders could make their parents proud.Y COMBINATOR
On a Tuesday afternoon in February, a cohort of startup founders, guided by Garry Tan, their investor and mentor (known as a “group partner” in YC terminology), gather in the original Mountain View building of Y Combinator for office hours. Seated in a circle around Tan and four other YC group partners, the 17 founders (all male) take turns presenting the latest two-sentence pitch of their businesses, along with the unique insight driving their work and a status report. Tan and his colleagues, seasoned in hearing thousands of pitches, effortlessly deconstruct and reconstruct them for improvement.
“If the number is accurate, just state it,” advises Tan to one founder who hesitated to quantify his market opportunity for fear of appearing presumptuous. Tan warns another about the importance of clarity in describing why people are using their tools, emphasizing that a sentence about traction should focus on being better, faster, or cheaper. When a third founder expresses difficulty in closing a crucial trial customer in Atlanta, group partner Brad Flora suggests hopping on a plane that very night—an initially perceived joke turned serious advice. “Do some LinkedIn research and engineer it socially so you’re there,” Tan suggests. “Put in the effort and make the trip in person. It makes a difference.” The other partners, all former founders, nod in agreement. “At this stage, they’re investing in you as much as the product,” adds Flora.
Tan views these exchanges as YC operating at its pinnacle. “We’re teaching engineers the art of selling,” he explains later. Taking individuals mostly rooted in technical backgrounds with ambitious ideas, YC educates them on the fundamentals of company-building over three months. “Am I focusing on the right objectives? What do customers desire? How do I collaborate with my cofounder? How do I acquire users? It’s fundamental,” explains Caldwell. “I’ve funded 1,000 companies. It’s like I’m a large language model trained on all this data.”
Handling two batches of 25 companies annually per partner is no small feat. With applications soaring into the thousands, Caldwell initiated an admissions team in 2015 to screen and interview applicants. By 2021, the team had grown to five members and was leveraging software to analyze founder responses to handle the influx of over 30,000 applicants. Despite the challenges posed by the COVID lockdown, YC admitted a record 748 companies in 2021, each receiving $125,000, a reduction from the pre-pandemic $150,000.
This represented a shift in direction. In 2020, Seibel, then CEO of the accelerator unit and now managing director overseeing the batch, aimed to transform YC into a large and successful state university system rather than an Ivy League network. A fully remote YC could support more international companies and bypass constraints like physical space. Ralston concluded 2021 envisioning a future with 1,000 companies per batch, provided quality control could be maintained.
However, concerns arose about the quality of YC’s offerings during this time, marked by the transition of its in-person Demo Day to an online platform. “The brand suffered a bit. People felt the quality wasn’t as dense,” noted an investor with stakes in several YC startups. “Those were tough times,” remarked a YC founder from the current batch, who had previously completed the program twice. According to a former employee, “PG felt we’d strayed too far off the path.”
Was the bloating of batch sizes the ailment or merely a symptom? Ralston informed Forbes via email that batch sizes had already been reduced during his final year in charge, in 2022. Streamlining the batch process, from partner roles in admissions to organization and follow-up with founders afterward, was a key priority for Tan upon assuming leadership. Tan introduced several significant changes: group partners would resume leading admissions, with the same partner guiding a company throughout its journey from the initial interview to a potential IPO. Additionally, YC would scale back its support for later-stage companies that had already secured funding and begun generating revenue, as well as non-software ventures.
Within the batch, Tan introduced Launch Bookface Live, a Friday night demonstration session for founders looking to softly launch a product. These reveals, a longstanding tradition, were previously limited to Bookface, the software directory and message board for YC alumni. The updated version includes a celebratory outing to a nearby establishment afterward.
To foster a more intimate environment within larger batches, Tan formalized “The Sharding,” inspired by database engineering principles, which organizes YC’s groups into parallel tracks. Groups, especially during the remote years, spent more time together than with others. By formalizing “sharding,” the founders are organized into tracks, with dedicated individual dinners and events, enabling Pinterest cofounder Ben Silbermann to engage with founders on one floor while GitLab CEO Sid Sijbrandij interacts with others on a different level.
The most recent batch attracted a record 27,000 applicants. Among the 260 accepted companies, founders generally expressed satisfaction with their halfway point experiences. Ka Ling Wu, cofounder and CEO of analytics startup Upsolve AI, relocated temporarily from London to participate in the program and now contemplates staying in San Francisco after finding camaraderie among its founders. Umur Cubukcu, who sold his previous company to Microsoft and has served as a YC visiting partner twice, returned with cloud infrastructure startup Ubicloud. He asserts that the program remains the optimal environment for founders to collaboratively develop an early product, despite the unchanged fundamentals of business formation. Many YC founders can turn to their batchmates for initial customer support, presenting significant potential among partners and alumni alike.
While AI founders originating from Anthropic, Google, or OpenAI can secure millions before venturing out on their own, they are unlikely to apply, even to the revamped YC. Yet, such “obvious founders” have never comprised YC’s primary clientele, according to Liz Wessel, a partner at First Round, founder alum, and recent YC visiting partner. By refocusing on earlier-stage startups, potentially younger in nature, YC is attracting founders of similar quality to its inception, she asserts.
With Tan’s adjustments implemented, YC may be tempted to scale up its numbers once again. According to Chesky, YC is already seeking additional group partners, which could increase capacity for dozens more participants. Caldwell dismisses the perennial debate over batch size as “tedious,” emphasizing YC’s lack of a fixed formula for capping its batches. However, some, including Graham, express regret over the program’s current record-low acceptance rate. “You should accept any sufficiently good startup,” he suggests. “If you lack enough partners, they’ll just have to work harder.”
Tan acknowledges the allure of Metcalfe’s Law, which posits that the larger YC’s network, the better. Nevertheless, Tan rejects the notion outright, evoking Seibel’s analogy of a state university system versus an Ivy League network. “This isn’t McDonald’s. This is actually The French Laundry,” he quips, referencing the exclusive Napa Valley restaurant of seven Michelin star-holding chef Thomas Keller. “What I’m trying to convey to each group partner is that there are 14 French Laundries, and each one represents its unique version of Thomas Keller.”
YC'S TOP TEN
Y Combinator’s alumni list still starts with Airbnb and Stripe. But more recent graduates like Brex and Deel are moving up its all-time highest-valued ranks.
| Company | Valuation ($ bil) | Year Attended | Public? |
|---|---|---|---|
| Airbnb | 104 | Winter 2009 | Yes |
| Stripe | 65 | Summer 2009 | No |
| Coinbase | 54.4 | Summer 2012 | Yes |
| DoorDash | 53.7 | Summer 2013 | Yes |
| Faire | 12.6 | Winter 2017 | No |
| Brex | 12.3 | Winter 2017 | No |
| Deel | 12 | Winter 2019 | No |
| GitLab | 11.5 | Winter 2015 | Yes |
| Rippling | 11.3 | Winter 2017 | No |
| Instacart | 9.4 | Summer 2012 | Yes |
Some alumni founders supported by Y Combinator’s growth fund are less enthusiastic about Tan’s notable changes. His actions to reprioritize the batch and centralize power around group partners contributed to the sudden layoff of several non-core staff members last year, including YC’s admissions, investing, and research teams.
Under pressure with the news looming, Tan followed Chesky’s advice to swiftly and decisively handle the cuts. In March 2023, Y Combinator’s growth fund Continuity and its related programs were terminated. Seventeen individuals, roughly 20% of the full-time staff, were laid off in what Tan described as a strategic shift, not a reflection of performance.
Ali Rowghani, the fund’s leader alongside partner Anu Hariharan, informed staff via Zoom, accompanied by a human resources representative. (Rowghani and Hariharan attempted to spin out the fund as a last resort, but the idea was swiftly dismissed as impractical, and the fund leaders departed.) Tan did not attend the Zoom meeting, and most employees never directly heard from him, as their email access was terminated before his company-wide announcement about the move.
The reaction of some of the approximately 20 companies that suddenly found themselves without a board director was more damaging to YC’s esteemed reputation. Founders from prominent recent alumni, including Brex, Deel, Monzo, Rappi, and Whatnot, signed a letter to YC’s board, denouncing the move and urging YC to reconsider or retain Rowghani and Hariharan on their boards. Their efforts were unsuccessful. Rowghani declined to comment, while Hariharan did not respond to a request for comment.
Tan acknowledged the dissatisfaction with the move. “I can understand their frustration,” Tan stated. “YC supports Anu and Ali in their future endeavors and anticipates ongoing collaboration with them.”
For some Continuity-backed founders, the handling of the move and its aftermath raised doubts about Tan’s ability to lead such a significant organization. “It felt like he underestimated the backlash or was pressured by someone else,” one founder remarked. The move was perceived by some as influenced by Graham’s guidance. Until 2022, YC operated with a board of overseers, known as BOO, which mainly left Altman and Ralston to their devices, according to insiders. However, Tan was Graham’s chosen leader, approved by a board consisting of him, Livingston, Levy, Chesky, and the outgoing CEO Ralston.
When questioned about these events, Tan assumed full responsibility. Graham and Livingston expressed support for Tan’s decision as a necessary shift for YC to diverge from behaving like a typical venture capital firm. “The questions that concern VCs are so mundane,” Graham stated. “If you enter that realm, you start to resemble those individuals, who are fundamentally different.”
Tan took to X, formerly Twitter, to celebrate as moderate candidates backed by non-profit GrowSF dominated party elections in San Francisco. As a board director at GrowSF, founded by Sachin Agarwal, Tan’s cofounder in his YC-backed startup 15 years ago, he had spent days sharing support for its efforts with his 422,000 followers. He amplified voices criticizing its opponents. “We are winning,” Tan declared alongside a photo of himself and Agarwal from 2009. “The boom loop begins.”
For those unfamiliar with the term “boom loop,” it represents the converse of a “doom loop,” a sarcastic term for San Francisco’s issues with crime, education, and housing. On one side are pro-“abundance” and growth-oriented “YIMBYs,” while on the other are San Francisco’s progressive “NIMBYs,” whom Tan and allies perceive as hindrances to city progress. These views overlap with “e/acc” (effective accelerationism), a recent movement Tan subscribes to, believing in technology’s capacity to reshape society and create abundance. Tan clarified that while some in the movement advocate for extremist totalitarianism and a robot-centric future, his stance favors humans.
Although Tan’s personal brand-building, especially on YouTube, played a significant role in raising Initialized, his VC firm’s profile, he is employing the same strategy across YC. “The more authentic you are on social media, the more effective it is,” Tan remarked. During its heyday, YC’s visibility grew through Graham’s popular essays and startup how-tos. Tan views video podcasts as YC’s new touchpoints, such as the Lightcone series, where Tan and three partners discuss emerging tech trends. “I have a brand, but I want there to be 14 PGs at YC,” Tan remarked.
However, Tan’s brand has not been without controversy. While his personal posts require only a few minutes per day, their impact can be significant. For instance, he once deleted a message apologizing for telling local political opponents to “die slow motherfuckers,” quoting Tupac lyrics. Another incident involved Tan accusing the leader of another accelerator called Neo of “slander,” leading to a public dispute on X.
These controversies have given some in the YC community pause. Several no longer feel comfortable expressing dissenting views on Bookface or X, where they risk mass downvotes from Tan loyalists. Tan blocked an alumni founder from Gaza who questioned his resharing of what they deemed pro-Israel, anti-Palestinian posts (he later unblocked him). A request for Y Combinator to express similar sentiment for the Palestinian community went unanswered.
“I don’t think Garry being vocal on such matters benefits the brand or the YC name,” one founder expressed. “I remain a strong supporter of YC,” added another. “But ideally, Garry would set politics aside and focus on improving YC.”
While Tan acknowledged the feedback, he has no immediate plans to change his approach. He maintains that he is not driven by personal fame or aspirations for public office, which he believes would strain his marriage. “I don’t want it to be about me, but if necessary, I’ll do it,” Tan affirmed. “I’m committed to the cause.”