十分钟看懂创始人：Jessica Livingston 的社交雷达

Lenny Rachitsky opens with Paul Graham's line that if you do not know Jessica Livingston, you do not understand YC. Livingston says she is often left out of news articles and that Wikipedia editors have tried to remove her entry. She shrugs it off, saying the people who matter in YC and Silicon Valley know what she did, though she suspects some people prefer a narrative with no woman on the founding team.

Her nickname came from the fact that her three co-founders, Paul Graham, Robert Morris and Trevor Blackwell, were deeply technical and she was not. In the early ten-minute interviews she kept time and mostly stayed quiet, watching whether the founders got along, whether they were committed and whether they really planned to quit their jobs. Afterward the others would ask her whether to fund them.

She says Graham in particular could get so excited about an idea that he started coaching the founders mid-interview. Livingston would pull him back with observations, such as one founder cutting off the other or two founders who worked at Google and never said they would quit. She says the group almost always agreed in the end.

Asked about successful founders who are difficult, Livingston says she would probably have funded Steve Jobs because she is persuaded by people who know their domain and love their product. But early on she consciously said no to people she thought were jerks, sometimes noting that she could not imagine having dinner with them every week. She says YC has never regretted those calls.

That filter became the basis of the YC community, which she says now includes thousands of founders. On Bookface, the forum for founders, someone will usually answer a question, and she traces that to an early culture of older batches helping newer ones and paying it forward.

She also reviewed applications for odd signals, a job that software now does by flagging them for readers. Examples include an equity split like 99% and 1%, founders who may not quit or relocate, or a company already 90% owned by someone else. None of these disqualifies anyone, she says, but each is a prompt to ask.

In interviews, Livingston says, Graham, Morris and Blackwell would grill the founders on their idea. The best founders answered something like: I have thought about that, and here is what I figured out. Defensive ones closed up as though they were being interrogated, which she calls always a bad sign. She also got excited by domain expertise and by founders fixing their own problem in a broken industry.

She explains why this predicts success. Startup founders are constantly questioned by skeptics they have to educate, and the first idea is rarely right. She cites PayPal, which began with money transfer between Palm Pilots until users pushed for a web version, and says founders must be open-minded enough to see what users actually want.

The best founders also want to learn from others, she says, and uses the Collison brothers as an example of people who listen to your answers when they ask a question. She also describes warning signs in co-founder relationships, such as one founder putting an arm out to answer for the other, or a programmer in a cage with a token amount of equity and no say in the product.

Livingston recalls the Airbnb interview during the winter 2009 batch, after the market crashed, when YC was funding only founders it thought could live like cockroaches. She says the team, Brian, Joe and Nate, had contagious energy even though she and Graham disliked the idea of staying in someone's airbed, and Graham tried to get them to change it. Then Joe produced the cereal boxes, Obama O's and Cap'n McCain's, which they had refilled with off-brand cereal and glued shut.

She thought they would work hard to do whatever it took, and she says it was clear they were the right bet. Rachitsky, who worked at Airbnb, says the story matches how the company tells it. Livingston adds that Graham called to accept them while they were driving and lost reception, and she thinks Joe pulled out the boxes on the way out, with some reluctance from Nate.

The other time she says she pushed the group to fund someone was GOAT. The founders, Eddie and Daishin, applied with an idea for group dinners with strangers, but they had previously run a cream puff company and told stories about how hard it was. She saw hustlers, and although that idea failed, they pivoted into the sneaker business. She also mentions Parker Conrad, who applied with Zenefits, unsexy HR and benefits work, because he knew how broken it was.

Livingston dislikes the word charisma because plenty of charismatic people are full of baloney and she does not want anyone fooled by it. She says it helps, since founders must sell, recruit and persuade investors, but there has to be substance behind it. Some uncharismatic founders are excellent if they know what they are doing and care about their users.

The trait she values most, she says, is earnestness, which she groups with authenticity. Earnest founders say honestly when they do not know an answer and explain how they would find it, instead of using smoke and mirrors. She contrasts them with a group of 45-year-old men building a fashion app for teenagers who clearly chose it for easy money.

Confidence also matters, and she says it is the ability to admit a gap without getting defensive. She adds that YC helps founders gain confidence in the three months after acceptance by helping them build a startup that is truly a good bet, so that on demo day they can say why.

Livingston says people do not know how hard a startup is until they do one. Starting on the side is fine, but at some point founders must burn the boat, because those who kept a paycheck and insurance did not quit their jobs when it got tough, and co-founders who stayed in their jobs often ended up leaving. She accepts that some people cannot leave for financial reasons, but they should not start a company unless they are prepared to.

On co-founders, she says she spent much of her time at YC mediating disputes, and many startups died or nearly died because founders broke up. In interviews she searched for any sign they did not get along, and liked to see them finish each other's sentences. She loves founders with a shared history, such as school friends, former colleagues or siblings, and calls two strangers getting together to start a company a massive red flag, though she notes Dropbox worked out.

She adds relentlessly resourceful, Graham's phrase, to the list. When Rachitsky says he avoids pre-seed because it is so hard, she explains that there is very little data when someone applies with only an idea, so a founder's past projects and schools are among the best predictors.

Graham used to call her Detective Livingston, she says, because she cannot help noticing small things, like a changed T-shirt, and wanting to know why. She links it to a childhood spent dissecting social situations with friends, and says she has always disliked phony people. She cannot say whether it is genetic or learned.

She does not train it deliberately but checks her gut calls years later. Sometimes it takes five to ten years before a founder she disliked fails, and she is glad to report that nobody she strongly opposed funding has gone on to be a big success. She says the impression is more intellectual than visceral, driven by a natural interest in people.

She has also been fooled. She describes Ilya Lichtenstein of MixRank, who she says was later convicted of stealing billions of dollars in Bitcoin, and she did not see that coming. More commonly, she says, YC accepts people who seem impressive in a ten-minute interview and then turn out to be limp or posers, or does not notice two founders who do not get along until one disappears a week into the program.

Asked how others can build the skill, Livingston says YC's advantage was partly that her co-founders were technical when many investors then were not. Her advice is to pay attention to subtle cues, because there are no trick questions. Keep a mental checklist: do the founders seem to get along, do they understand their product, and are they defensive when asked why they beat a competitor?

She suggests simply asking how the founders met, whether they have worked together before, and how it is going. Afterward, take time to reflect on how the conversation went. Rachitsky points to the Reading the Mind in the Eyes quiz, on which she scored 36 out of 36 after Graham posted his 25. The best Rachitsky's friends managed was 31.

Her method for the quiz was to look at the eyes and ask what the person was trying to communicate. She also noted that understanding the nuance of each emotion word matters, such as irritated versus angry. After Graham tweeted her score, about ten people at a YC event in London asked for her secret.

Livingston started The Social Radars, co-hosted with YC lawyer Carolynn Levy, because she lives in the English countryside and missed catching up with YC alumni. She was inspired by the actor podcast SmartLess, where one host invites a guest the others do not know, which produces genuine conversation. She wanted something similarly unscripted for startup founders, and says guests have included Paul Graham, Patrick and John Collison, Brian Chesky and Emmett Shear.

Her main lesson as an interviewer is that she talks too much, and listening back she hears herself ask a question in twenty sentences when two would do. She says hosts should speak often but briefly and sometimes interrupt long answers. Guests also open up because they trust her, since she promises they can review the episode before release and is not hunting for gotcha moments.

The episode that sticks with her most is Parker Conrad's. She says it covered his Zenefits experience, including a smear campaign and legal threats she says came from someone still at the company, and that no reporter ever corrected the record. She says she cares about justice and was glad to give him a place to tell what happened. On metrics, she says that as long as she has twenty interested listeners she will keep going, while Rachitsky suggests watching downloads and subscribers.

Looking back, Livingston calls the early days magical and pure. Nobody knew about YC, only eight founders joined the first summer, and no one was scrutinizing or writing negative stories about the group. She is more sensitive than Graham to untrue press, which is part of why she remembers the early days as the most fun and productive time of her life. She is now thinking about writing a YC biography.

She names two moments when she felt YC was working. The first was after the first Summer Founders Program, whose batch included Sam Altman, Emmett Shear, Justin Kan and the Reddit founders. The founders became friends, had weekly dinners with speakers and met investors on demo day, so the team decided to keep going. The second was about a year later, when Reddit was bought by Condé Nast and the press began to take notice.

Rachitsky highlights that the batch structure came from the founders having no experience as angel investors and wanting to learn by funding many startups at once. Livingston agrees, saying they planned to move to investing one at a time, but realized there was something special about the batch. Her wry conclusion is that she loves domain experts, but sometimes ignorance is bliss.