Blog

• 2026-11-02How Well Do Agents Perform on DevOps Tasks?Frontier agents on CloudOpsBench have the knowledge cloud tasks require and fail mainly at judging when a task is done. A three-step taxonomy of where they fail.
• 2026-10-13What cloud operations are, and what we testThe day-to-day work of a DevOps engineer, with examples from the task set, and how a verifier decides whether an agent did it.