Sunny

← All weeks

What I’ve been reading (W17, 2025)

Visions of Greenoaks, how WhatsApp thinks about building solutions, and what Buenos Aires is like

  1. Visions of Neil Mehta, founder/investor at Greenoaks https://joincolossus.com/article/the-visions-of-neil-mehta-greenoaks/
    • Run Greenoaks—a nine-person firm that has turned roughly $15 billion of committed capital across five funds and 55 core holdings into more than $13 billion of gross gains, compounding about 33 per cent for thirteen years.
    • As a young investor, he outbid a lone farmer for a Guangzhou apartment block, wandered BYD’s cluttered factory in 2007, and watched an entire Olympic grandstand tapping away on QQ and all reminders that the best edges often hide in odd corners if you’re willing to look.
    • Their math is simple: find a founder who feels like an artist—remarkable intellect, obsessive customer focus, unreasonable grit, clear ambition, and a willingness to diverge—and then become that person’s most important partner for decades.
    • They earn conviction the hard way; Mehta still “gets dirt under his fingernails,” calling customers, competitors, and suppliers before the first founder meeting, and can wire half a billion dollars inside 36 hours if the research lines up.
    • They see beauty not just in products but in the P&L itself - clean unit economics, good working-capital loop, or a business that breaks a trade-off to create a jaw-dropping customer experience (JDCE) is, to them, aesthetic art.
    • He has backed many with audacious checks: buying 40 percent of his debut fund on Coupang and later adding a $500 million rescue that returned more than 10x in Rippling; leading Wiz at a $1.7 billion valuation when revenue was only $2 million; wiring $500 million to Rippling overnight during the SVB panic; and accumulating nearly 4 percent of Carvana when the market priced it for bankruptcy.
    • Greenoaks avoids most capital-hungry AI labs, Mehta thinks of them as 1980s airlines with big capex and thin moats but also reportedly ready to write a $500-million-plus check to Ilya Sutskever’s secret Safe Superintelligence because Sutskever fits its “great artist” founder mould and is chasing genuine super-intelligence, not incremental model scaling.
  2. WhatsApp’s strategy for reliable messaging wasn’t just about altruism—it made good business sense. They aimed to reach every user on every device, whether an iPhone, BlackBerry, or a basic Nokia handset stuck on a slow 2G network. Engineers even tested the app in remote, signal-starved locations using old Nokia C3 phones to guarantee reliability under harsh conditions. Ignoring flashy features, WhatsApp focused solely on cheap, dependable texting, appealing especially to markets like India and Europe, where each SMS had a real cost. By designing for the lowest common denominator, they turned broad accessibility into their main distribution channel—if it worked in the worst conditions, it worked everywhere https://restofworld.org/2024/how-whatsapp-for-business-changed-the-world
  3. Buenos Aires is the meeting point between French facades, Italian food, and Latin passion, and it also has more soccer stadiums than any city in the world https://perell.com/essay/buenos-aires-the-paris-of-south-america
  4. Real growth comes from genuine value, not validation. Pay attention to signals you can’t fake—like when your product naturally integrates into your customers’ daily routines, teams start using it more because it addresses actual problems (not just because a sales pitch said it’d deliver ROI), or engineers spontaneously start building on your API without your sales team even knowing. https://writing.nikunjk.com/p/function-not-form
  5. Everything is converging on TikTok, which is TV with infinite channels https://rakhim.exotext.com/but-what-if-i-really-want-a-faster-horse