Actually Jessica does a lot more than "administrative duties" for YC. Of the four YC partners, she is the best judge of people, and that's the main thing we're judging at this early stage. It would not be too far from the truth to say that interviews consist of me and Rtm and Trevor asking founders a lot of questions, and then as soon as they walk out, turning to Jessica and asking "ok, should we fund them?"

> Of the four YC partners, she is the best judge of people, and that's the main thing we're judging at this early stage.

"Best judge of people" is sufficiently nonspecific to not be a testable claim.

If you mean she is best at judging who would be the most successful if you fund them, you lack the evidence to make that claim, as you don't know what would have happened with all the people you didn't fund, which is almost all of them.

If you mean that the ones that she picked -- but you didn't -- performed better than the ones you picked -- but she didn't -- it's possible that she is valuable not because she is so accurate, but that you are so inaccurate.

The notion that YC is especially good at choosing successes is not borne out by your results, or you'd all be a lot richer. And as mentioned, you don't know which were the ones that "got away" -- founders that went back to school because they didn't get in, but would have been even larger successes. Reddit would have been in that group if you hadn't had a last-minute change of heart.

As for the relentless founders, was it really that difficult to recognize that Sam Altman was one of those people? For the obvious cases, it's apparent to all four of you; and for the less-obvious cases, you lack the data to know how many were false negatives. All you know is the false positives; but that's going to be the most common outcome anyway.

Being ramen profitable working on something you don't especially like is silly compared to what you COULD be making at jobs in the Bay Area. A website is not a startup, and most of them lack any major potential, barring irrational market exuberance.

Considering Yahoo! Paid 6 billion for Broadcast.com and 4 billion for Geocities shortly after paying 95 million for ViaWeb just shows the most important skill is finding someone with billions of dollars and talking them into throwing it away on your unprofitable business.

Have you read Founders at Work? The interviewing is exceptionally good. It feels like the founders are just telling their stories. JL's questions almost disappear, even though they're right there in the text. You don't achieve that level of invisibility as an interviewer without a deep understanding of what your interviewees are talking about.

There are probably other ways you're undercrediting her. I only mention this one because it's the one I have personal experience with.

In fact, when we started YC Jessica was the only one of us with venture capital experience (having worked at a major investment bank in Boston since before she met PG.) She had a lot to do with us getting into the business, and her judgments about whom to fund have proven to be accurate.

Sorry to disappoint you by not being an all-girl seed funding firm with thigh-high pink boots.

> and her judgments about whom to fund have proven to be accurate.

As mentioned to Paul, you lack the data to establish what would have happened if you had funded other people instead.

> Sorry to disappoint you by not being an all-girl seed funding firm with thigh-high pink boots.

Start-up, not investment group. Referencing the lack of technical female founders, as mentioned in the article.

What you guys do is obviously not technical in the sense we were talking about if JL is better than the other three of you.

You have no idea what you are talking about. For one thing, from what I understand Jessica was the main force in starting up YC. No Jessica Livingston, no Y Combinator.

Tell Henry Ford and Ray Kroc that there's nothing exceptional about being the first to figure out how to organize something. And YC's impact might just end up being more than either of those.