Paul Graham
Paul Graham is a programmer, essayist, and co-founder of Y Combinator (2005) — the seed accelerator that funded Dropbox, Airbnb, Stripe, Reddit, and hundreds of others. Before YC he co-founded Viaweb (one of the first web application companies, sold to Yahoo in 1998 for ~$49M) and wrote On Lisp and ANSI Common Lisp. He stepped back from YC's day-to-day operations around 2014 and has published essays prolifically since the early 2000s at paulgraham.com.
His essays are unusually direct about contrarian observations — things that are obviously true once stated but that most people don't say. They cover startups, hackers, writing, and epistemics. Dylan Bowman's reading list for generalists draws heavily from Graham precisely because his essays teach disposition and judgment, not just tactics.
Recurring themes across ingested essays
Agency and resourcefulness — the single most common theme. Good founders are "relentlessly resourceful": not merely determined, but adaptable in the face of novel obstacles. Pure determination fails; you must keep trying new things. This shows up in relentlessly-resourceful, do-things-that-dont-scale, and what-we-look-for-in-founders.
Identity as cognitive constraint — keeping a small identity is a prerequisite for clear thinking. The more labels you attach to yourself, the dumber those labels make you. Politics and religion are common traps; so is "being a programmer" or "being a writer" if treated as fixed identity rather than practice. See keep-your-identity-small.
Epistemic honesty in argument — his disagreement hierarchy (DH0 name-calling through DH6 refuting the central point) is a practical tool for both critical reading and better writing. Real refutation requires committing to the central point and showing why it fails. See disagreement-hierarchy.
Essay as exploration — real essays are different from school essays. They start from questions, not theses. The word essai means "to try." Writing is how you form thoughts, not just record them. See essay-as-exploration, meander-principle, surprise-as-epistemic-target.
Unscalable early effort — startups don't take off on their own; founders have to crank them manually. The "Collison installation" (Stripe) and Airbnb's door-to-door listings are canonical examples. See do-things-that-dont-scale.
Connections
- relentlessly-resourceful
- do-things-that-dont-scale
- founder-qualities-paul-graham
- keep-your-identity-small
- disagreement-hierarchy
- essay-as-exploration
- meander-principle
- surprise-as-epistemic-target