I've been hearing "cluely" everywhere, or at least on a bunch of infosec podcasts about various insane bugs and exploits in their software. And when I heard the description of it, "an AI assistant that sits in meetings with you," and the emphasis on "nobody knows it's there," with the subtle wink-wink in the marketing... I was like, "hang on... this sounds a lot like the guy who got expelled from Columbia for making an app to do this in his Amazon interview..."

And now I realize it's the same guy! I don't know if he intended this from the beginning — he probably didn't, considering nobody forced Amazon to email Columbia and advocate for his expulsion. And maybe he wasn't even expelled. But damn. Love him or hate him, this is a masterclass in viral marketing and playing the hand you're dealt. Good for this dude. I wish him the best but personally the whole concept of a silent meeting participant gives me the creeps, even before considering the ethics of the "interview cheating" use case.

Roy, if you're reading this... if you succeed it will not be due to your engineering skill. Focus on marketing and put some funding towards engineering brains who can make a robust product that passes enterprise muster. Build some corporate growth hacks into the core product (e.g. SSO + calendar invite pathogen + Figma trick of inviting anyone with same corporate email and charging per seat in a post-facto "true up" at end of billing period). Then go hunting for sales people who have no ethical qualms with silent meeting participants taking notes and helping them look less clueless to their prospects.

And this post shows exactly why cluely is such a damp squib destined to fail. Zero substance just grandiose claims that hold no water. They’re spending tens of millions of dollars of investor money “going viral” instead of spending investor money actually building a great product.

Cluely is just some kids larping as founders, the perfect example of the worst of the AI boom, it’s as if the entire thing is AI generated, a simulacrum of a startup. There are a million more talented and capable teams of young people out there.

If someone from a16z reads roy’s post they might finally realize they should ask for their money back.

i'm a cluely hater but the fundamentals of the article are pretty good. simply, "viral content" is and ad. and ads need eyeballs and conversions.

whether they end up actually building a good product that isn't just cheating will determine if they will be seen as clowns or geniuses.