Harvey Co-founder: Own Your Intelligence Stack — Custom Models, Harnesses and Evals
dair_ai · x · 2026-09-23
Harvey co-founder Gabe Pereyra reflects on the hardest part of building Harvey: doing right by customers under pressure to do the easy thing.
- The easy path would have been forcing customers onto consumption pricing early and serving worse models to protect margins. Harvey refused.
- Instead it let customers transition on their own timeline while getting the best models, even at margin cost — optimizing via routing, harness improvements, and post-training to serve frontier intelligence affordably, plus usage dashboards and per-seat spend management infrastructure.
dair-ai's omarsar0 draws the broader lesson: 'Own your intelligence stack' — you can't scale to the frontier by renting intelligence; custom models, harnesses, and evals are becoming huge assets.
Related event: Harvey CEO explains path to positive gross margin(2 posts)→
More from coding & agent
- Interactions Tool Searches Web Animations by Motion for Claude Code to Rebuild — yuwen_lu_ · 2026-11-12
- Agent Substrate Calculator Estimates Session Resume Speed Under Load — pjausovec · 2026-11-12
- Senate Bill Would Force Big Platforms to Accept AI Agents, Raising Auth Privacy Questions — Prof_CarlaReyes · 2026-11-12
- Dell expands AI Data Platform: agents have a data-context problem, not a model problem — DavidLinthicum · 2026-11-12
- LLM controls vocal synth to perform Ave Maria in 4-part harmony — AAAzzam · 2026-11-12
- ArcGrav: a cross-model Agent Control Language to cut token overhead — ArchitecturalGravity · 2026-11-12