GitLab CPMO: code is abundant, trust is the bottleneck for AI coding
TechNadu · x · 2026-11-05
GitLab Chief Product and Marketing Officer Manav Khurana argues "code is abundant, trust is the bottleneck" — as agentic code generation grows, the hard part shifts from writing code to deciding if it's correct, secure, and shippable. Key points:
- Teams run a "shadow software factory": coding, issue tracking, CI/CD, security and deployment sit in separate tools.
- Agents may run on personal laptops with local credentials and unmanaged tools, each developer setting their own orchestration rules.
- Without a shared evidence chain from plan to production, engineers piece together what changed and what was approved.
His answer: a governed software factory with shared context, consistent controls, traceable agent actions, permission boundaries, security checks and human approvals before production — plus measurable ROI for leadership.
More from coding & agent
- Tencent's Hy4 lands on Vercel AI Gateway at 50% off through end of 2026 — TencentHunyuan · 2026-11-05
- Openbar puts Claude Code and Codex sessions in the Mac notch — approve diffs with ⌘Y — Significant-Sea-7238 · 2026-11-05
- "I could uninstall every terminal and be fine": dev says AI coding agents have replaced the command line for him — ethanniser · 2026-11-05
- Dev builds signed audit log for AI agents after prompt-injection test blocks $99,999 refund attack — Suicidal_Squirell166 · 2026-11-05
- ECC, a 297k-star harness packing 71 agents and 314 skills for Claude Code and beyond — victor_explore · 2026-11-05
- Dev builds terminal harness SUTRA for models, DeepSeek session costs just $0.0001 — cneuralnetwork · 2026-11-05