InfoQ - Automation

DoorDash Uses Multi Agent LLMs to Clean up 60,000 Feature Flags

Leela Kumili — Fri, 18 Sep 2026 13:50:00 GMT

DoorDash built a multi-agent LLM system to automate stale feature flag cleanup across more than 60,000 flags and 623 repositories. The workflow combines live experimentation data through MCP, engineer approval, isolated Git worktrees, parallel agents, and automated validation. In an evaluation of 50 flags, 45 produced usable pull requests at an average of 13.8 minutes and $4.79 per cleanup.

By Leela Kumili

Presentation: Teaching Engineers, Trusting AI: How Education Enabled Autonomous Code Review

Sarah Deitke — Wed, 16 Sep 2026 11:00:00 GMT

Sarah Deitke discusses how Duolingo drives cultural AI adoption beyond tooling access. She explains their internal AI literacy workshops and observability dashboards, then shares a case study on redesigning code review using an automated PR risk-assessment bot. Deitke demonstrates how pairing targeted developer education with safe AI guardrails speeds up delivery without increasing defect rates.

By Sarah Deitke

Article: Implementing Durable Workflows on Postgres Without an External Orchestrator

Raman Varma — Mon, 14 Sep 2026 11:00:00 GMT

Postgres can serve as the durable state store and coordination layer for workflows, eliminating the need for an external orchestrator. SKIP LOCKED enables concurrent work processing, primary-key checkpoints enforce idempotency, and leases support crash recovery. Workflow sleeps and human approvals can also be persisted as database state and survive restarts.

By Raman Varma