← Back to context Comment by senordevnyc 6 hours ago How does this compare to something like Cursor Cloud Agents with a solid set of skills and tools? 1 comment senordevnyc Reply danoandco 6 hours ago Similar but reusing lab-native CLIs like Claude Code or Codex, which they perform RL on. And so in the long-run, we believe this approach wins over custom harnesses.
danoandco 6 hours ago Similar but reusing lab-native CLIs like Claude Code or Codex, which they perform RL on. And so in the long-run, we believe this approach wins over custom harnesses.
Similar but reusing lab-native CLIs like Claude Code or Codex, which they perform RL on. And so in the long-run, we believe this approach wins over custom harnesses.