Create an account to save sources and durable progress.
Lenny's Newsletter · Claire Vo
podcastplaybookSource excerpt
About Lenny's NewsletterWatch now (46 mins) | 🎙️ Educator and developer John Lindquist shares eight Jev demos, from live voice classification and deduplication to agent routing, showing how low costs make once-impractical ideas worth trying
This is a limited feed-provided excerpt, not the full original work.
Apollo layer
A major drop in AI cost and latency can make previously uneconomic product ideas viable. Revisit workflows such as real-time classification, deduplication, and agent routing, then test whether faster, cheaper inference creates meaningful customer value rather than merely a novel demo.
Model economics shape product strategy. Lower per-task costs can support higher-frequency usage, real-time experiences, and experiments that were previously too expensive—potentially changing what you can offer and how you price it.
List three customer workflows you previously rejected because AI was too slow or costly. For each, define a one-week prototype, the customer outcome it must improve, and the maximum inference cost that would support a sustainable margin.
Private notes
© Apollo Advisors° ABN 34 346 108 139