Keeping Humans Accountable as AI Agents Take On the SDLC
DevOps.com, Monday, September 28th, 2026
Atlassian's Ming Wu says governed agent loops can automate SDLC work, but people must stay accountable for results.
In a DevOps.com interview with Alan Shimel, Ming Wu, head of engineering for Dev AI at Atlassian, argues that responsibility for agent-completed development tasks stays with people.
Wu describes governed agent loops, in which governance provides visibility, guardrails and enforcement while loops let agents process batches of work such as taking an issue, doing the work and reviewing a pull request.
Jira serves as the initial interface, with support for third-party coding agents and a shared context layer for priorities and business intent, as customers look to scale these workflows and prove real delivery gains.