Y Combinator, the startup accelerator known for discovering first-time entrepreneurs, has quietly seen a surge in founders who return for a second run through its program. An analysis by Crunchbase News of data provided directly by YC revealed that 454 repeat founders have participated in the accelerator's cohorts, generating 935 founder-company records between 2005 and 2026. The pattern shows that coming back to YC has largely been a two-act story for most entrepreneurs.
The numbers paint a picture of calculated returns and occasional rapid-fire restarts. Of the 454 repeat participants, 428 founders—94% of the total—went through YC exactly twice, while just 25 appeared three times. Twitch and Stash co-founder Justin Kan stood alone as the only four-time participant. Founders typically waited about five years before rejoining, with an average gap of 5.1 years separating their appearances. But the data splits into two camps: nearly 30% of return visits happened within two years, including 38 founders who came back within the same calendar year, while 61 returns occurred after a decade or longer. Repeat founder numbers climbed to 65 in 2025, though the report notes this peak coincides with significantly larger cohort sizes in recent years and potentially incomplete 2025-26 data. Some founding teams even returned intact, including those behind Layer by Layer, Voodoo Manufacturing, Ultra, Blair, and Fastgen.
Aaron Epstein, a general partner at YC who worked the spring 2026 batch, said the trend "definitely feels like more" now, though he's quick to point out it's not entirely new—the growing alumni pool naturally creates more eligible returners. According to Epstein, who has worked with over 1,000 startups at the accelerator and previously co-founded Creative Market (YC W10), second-time founders "know exactly how to get the most out of the advice, network and resources" and excel at separating signal from noise. The biggest mistake they avoid, he noted, is "overhiring or overspending pre-product-market fit"—a regret he sees among successful first-timers who hired too many people, slowed down, and stopped enjoying their own companies. Sherwood Callaway, founder and CEO of Sazabi, embodies this pattern: after his first company Opkit (an insurance verification startup later acquired by 11x AI) felt like "an MBA case study approach," he returned with Sazabi, an AI-native observability platform competing with Datadog that raised $8 million in seed funding in late June 2026 from J2 Ventures, Village Global, and YC.
Epstein believes AI is accelerating a shift toward leaner teams, much like cloud computing eliminated the need to raise large sums for servers—making it possible for solo founders with strong networks to "produce at 10x or 100x what a traditional engineer would be able to build." Many repeat participants are starting solo but leveraging existing relationships to bring in founding employees, moving faster while avoiding the loneliness of building alone. Callaway himself was more strategic the second time, deferring his batch to build out more product first and using YC as "a go-to-market acceleration event"—a tactic he wouldn't have known without prior experience. He also noticed a more experienced spring 2026 cohort and new anxieties around durable moats "in an AI world when lines of code are effectively free," though he compared the fundraising energy to fall 2021—this time driven by AI and real material gains rather than zero interest rates.
As YC's alumni base continues to expand, the accelerator expects more second-time founders to appear in future batches, drawn by personalized partner advice, a community of ambitious peers, investor access, and the urgency of the batch environment. Callaway's thesis for Sazabi captures the moment: "AI has changed how software gets written. Now it is changing how software gets operated." For accelerators competing with YC, the repeat-founder phenomenon presents a retention challenge that goes beyond initial selection—alumni loyalty becomes a competitive moat in itself. Programs that can't offer comparable long-term value may find their best graduates looking elsewhere when they're ready to build again.