About this session
Most agents handle one prompt and a quick burst of work. Hyperagent runs agents that wake on a schedule, work for hours in real browsers and code environments, and escalate to humans. Built on Vercel Sandbox with AI Gateway, we tested 14 models on 42 real jobs and found a 50x cost gap on identical tasks.
What you’ll take away
- Explore scheduled, long-running agent execution in browsers and code environments.
- Understand where human escalation belongs in an autonomous workflow.
- Compare model cost using real tasks rather than headline pricing.
Speaker
Brian O'Rourke
Founding Team · Hyperagent
Brian O'Rourke is part of the founding team at Hyperagent. At Ship, he explores agents that work for hours rather than seconds: running browsers and code environments, waking on a schedule, and knowing when to involve a human. His session draws on comparisons of 14 models across 42 real jobs.