Y Combinator co-founder Jessica Livingston on the San Francisco comeback, founder mode, and the carpetbagger problem

Summary

Jessica Livingston, co-founder of Y Combinator and current board member, offers a notably optimistic read on San Francisco's recovery. Having moved to England in 2016 and returning only two to three times annually, she describes the Bay Area today as "vibrant and alive" — a stark contrast to a visit roughly three years ago when YC portfolio companies including Reddit and Twitch were empty, staff still working remotely. Her outsider cadence gives the observation some credibility: she is not a daily participant rationalising sunk costs.

YC itself has expanded to four batches per year, up from two during its earlier years. Livingston sits on the board but is not operationally active, a deliberate arrangement she attributes to burnout from simultaneously founding the organisation and raising children with minimal maternity leave. The distance, she argues, preserves her enthusiasm rather than diminishing her utility.

YC's Origin Logic and Enduring Thesis

Livingston confirms that YC's founding insight in 2005 in Boston was structural, not philosophical. Early-stage capital simply did not exist in a usable form. The minimum viable raise was a few tens of thousands of dollars to cover rent and quit a day job, but VCs only wrote million-dollar cheques and required formal business plans. YC's early value proposition was cash, legal scaffolding, and business coaching targeted specifically at technical founders who could build but found incorporation paralyzing.

The organisation ran bi-coastal sessions — Cambridge summers, Mountain View winters — for three years before consolidating in Silicon Valley after Livingston and Paul Graham had their first child. The decision to stay geographically concentrated was deliberate and has not wavered. Requests to franchise the model to New York, London, or Boston were consistently declined. The thesis is that the Bay Area ecosystem is genuinely unreplicable, and requiring founders to be physically present for the batch — even if they return home afterward — is non-negotiable.

The first batch had eight companies. Reddit was the only one to demonstrate meaningful product-market fit, and even that was gradual.

The Founder Status Shift and the Carpetbagger Problem

Livingston does not believe the glamorisation of startup founding was predictable in 2005. The original YC cohort was narrow — technically exceptional people who did not fit conventional career paths. The Social Network, released around 2010-2011, produced a measurable spike in YC applications. Parents who had previously pressured children to choose Stanford graduate school over YC began to reconsider.

The status elevation has a downside Livingston flags directly. A segment of applicants now treats YC acceptance as a resume line, listing it under education on LinkedIn. YC has had to take active steps to prevent that framing. More broadly, she is sceptical of founders whose primary orientation is liquidity rather than product. Her term for them is carpetbaggers. Her practical argument against the model is not moral but operational: the difficulty and duration of building a company — she advises founders to commit at least a decade — means purely financial motivation tends to erode before an exit materialises.

On team size, she reinforces a position consistent with YC's documented advice. Staying small before product-market fit is confirmed preserves the ability to pivot. A company at 30 to 50 employees without genuine market pull faces structural inertia that makes a bet-the-company pivot extremely difficult. She frames this as a capital efficiency argument as much as an operational one.

Product-Market Fit and AI Revenue Inflation

On distinguishing real product-market fit from artificial signals, Livingston cites Paul Buchheit's formulation: it is better to have a small group of users who love the product than a large group who merely like it. Paying customers are the clearest signal. Friends who say they will use something are not a data point.

The question of AI inference reselling — effectively subsidising customer acquisition by selling compute below cost to generate revenue that reads as growth — is treated as a contemporary version of the same false signal problem. Livingston does not engage it technically, but her framework is applicable: if customers are consuming a product because it is priced below cost rather than because it solves a problem, the underlying business has not demonstrated love.

Founder Mode Series

Livingston launched the Founder Mode podcast series — co-hosted with Carolyn Levy, YC's legal counsel, on their existing Social Radars podcast — on the day of this interview. The series originated from a talk Brian Chesky gave at a YC retreat approximately a year prior, which Livingston describes as a defining account of how founders can be gaslit into delegating core decisions inappropriately. Paul Graham subsequently wrote an essay formalising the concept. The new series collects specific examples from YC founders. The launch batch includes episodes with Brian Chesky, Gary Tan, and Paul Graham, with roughly 10 total episodes to be released over the following weeks. The retreat that sourced the content was held in Sonoma Valley and organised by Gary Tan.

YC turns 20 years old in 2025, having been founded in 2005.