Skip to content

“is definitely an issue where if you have 10 million lines…” — Y Combinator Startup Podcast at 6:14

“is definitely an issue where if you have 10 million lines of code, that's maybe 100 million tokens. And both having a model that can actually ingest that and having it be cost-effective, and then not just having a model that can physically ingest that into its weights, but also one that actually pays attention…”

0:00 37:28