Rakuten fixes issues twice as fast with Codex

Rakuten has integrated OpenAI's Codex coding agent across its engineering stack, achieving ~50% reduction in mean time to recovery (MTTR) and compressing quarter-long projects into weeks.

Three Pillars of Rakuten's AI Strategy

Build Faster ("Speed!! Speed!! Speed!!")

Teams use Codex in operational workflows alongside KQL-based monitoring to accelerate root-cause analysis and remediation.

Build Safer ("Get things done")

Codex is invoked in CI/CD pipelines for code review and vulnerability checks, applying internal coding standards automatically.

Operate Smarter ("AI-nization")

Codex drives larger, ambiguous projects from specification toward working implementations, reducing dependence on perfectly-defined requirements.

50% Faster Incident Recovery

When issues occur, Codex helps identify root causes and suggest fixes by working alongside KQL monitoring workflows. Instead of manually stitching together queries, logs, and patches, engineers focus on validating and deploying fixes. Rakuten estimates ~50% reduction in MTTR.

CI/CD Integration

Rakuten feeds internal coding principles into Codex workflows so reviews align with company standards consistently and automatically. As shipping accelerates, automated review prevents review from becoming a bottleneck.

Full-Stack from Single Specs

For complex projects, Codex executes full-stack builds from specifications — reducing quarter-long efforts to weeks by enabling more autonomous execution.

"We don't just care about generating code quickly. We care about shipping safely. Speed without safety is not success." — Yusuke Kaji, GM of AI for Business at Rakuten

Source: OpenAI Blog