From the archives of posts which are basically reflections and that I might never post at all. Strongly relates to: To Optimize, Don’t Optimize, yet to be published.

If you want to meet and talk to cool people, your first thought might be… to try to meet cool people. Like, at a meetup or something, maybe? So then, why are the cool people so rarely at the obvious venues: founder hangouts, dating app queues, online meetups and events?

[Note: In the next few examples, I’m going to try to point at some vibes that I sometimes get. Obviously, there are exceptions to each of the following statements, and they’re only true to varying degrees. Confidence exaggerated for effect.]

Here’s a very cynical take: entrepreneurs at founder meetups in SF are almost never successful founders (or, most likely, even those to-be-successful), because if they were, they wouldn’t have time to go. People on dating apps have more flaws than the average, because if they were naturally appealing, they’d already be in a relationship.1 People who go to publicly-listed meetups aren’t good at socializing and hence lack a solid social circle, because if they were, they’d already have a social surplus and be overwhelmed with higher-signal unlisted events from friends. People who apply to online job applications are not people you want to hire, and so on.2 The act of trying in the most obvious manner is an anti-signal, because nobody who’s actually good needs to try like that. They have better backdoors.

Of course, this can’t be entirely true. Somehow, people start at the beginning and slowly meet cooler people. But I think it’s mostly true. In that case, how do people improve?

How do you escape social mediocrity?

Partial Answer 1: Be or become legibly better than everyone else in some measurable way. If you have 100% on LeetCode, score 170 IQ, and cure cancer, you’ll get noticed even if you pick the noisiest default channel to present yourself on. If you’re really attractive, you’ll get noticed just as you go through life.

Partial Answer 2: Make your own backdoor, by choosing not to directly optimize for the obvious thing (meeting people, dating, hiring, investing) and instead exposing yourself to unadulterated human potential through events orthogonal to what you really want. Instead of holding founder-VC dinners, run hackathons. Instead of going on dating apps, just meet people while rock climbing.

This lets you sample for whatever population you want, rather than the (probably lower quality) slice of people who are also desperately optimizing alongside you, at the cost of false positives, since not everyone who attends will have the same goals you do. This is a downside compared to the direct optimization approach. However, this also allows you to show other aspects of yourself, beyond those that are legibly targeted at the direct goal. You might not have a perfect LeetCode score, but if you befriend the Meta recruiter at the bar through witty commentary, you’ve effectively gotten a backdoor into being remembered.

Of course, do this too much and quite rapidly your VC-sponsored hackathon becomes a transparent recruiting event, as are most corporate events these days. Why is that? Why did optimizing hurt the very thing we wanted to achieve?

The economic viewpoint

Another view explaining the above: everything is a market, and markets follow the efficient market hypothesis. If it were possible to find or host a public event that Really Attracted Interesting People Every Time, it’d be swarmed and rapidly be filled by uninteresting people. So much like the market, all public alpha rapidly becomes worthless. The intrinsic difficulty of recruiting is perhaps to continuously locate nonobvious slices of the world that capture the demographics you want; it’s a delicate balance between optimization and over-exploitation.

But that doesn’t mean you can’t find gaps in the market: places where cool people congregate that aren’t known to most. And considering most people don’t explicitly optimize social dynamics vs. e.g. the stock market, there are probably a lot of gaps, if you can find them.3

This is why colleges, local neighborhoods, Twitter bubbles, workplaces, and any other assorted niches are good: the market is so small as to be entirely unoptimized. That’s why I can sneak into an EA party and suddenly be talking to HARM TO ONGOING MATTER. This is also why dating / making friends locally, or through school, is (or was) good.


  • Openness is inversely proportional to quality: Online events < public regional meetups < campus-wide activities < private exclusive events.
  • Formality is inversely proportional to quality (less formal = less optimized): Hiring fairs < arranged one-on-ones < coffee chats < actually spontaneous coffee chats between friends.
  • This is why flirting is socially preferred compared to asking someone out point-blank. By giving yourself plausible deniability, you’re taking yourself out of the class of desperate optimizer-y people who do the default action.

Unsolved problems

  • How does intrinsic motivation play into this? Why do people so often observe that increasing your own self confidence makes you more attractive (in all ways) to others?
  • How does availability play into this? How do you know that you’re in a rock climbing group with people actually open to new connections vs one where it’s only for climbing?
  • How do you pick places that actually have slices of humans you like? Essentially, how do you find your people?
  • Is it possible to optimize (select well on a large scale) while not destroying quality? Is there perhaps an automated way of vetting the masses, or is this really a thing that will forever remain only in the domain of ad-hoc, unoptimized human interaction?


  1. See Sasha Chapin

  2. A friend reports a similar example: You can’t find the mycology experts on You have to go through weird backchannels and random introductions via someone you met at a party that one time. 

  3. Hmm, what’s the index fund of social interaction? Going to church?