Coding Agents Move the Bottleneck to CI: SonarSource on Rebuilding Merge Queues for AI-Scale Commits
AI Engineer · youtube · 2026-10-14
At AI Engineer, Gautam Korlam (Uber's SubmitQueue, now Gitar at SonarSource) argues coding agents produce changes faster than the queues that test, review, and merge them, shifting the bottleneck to CI and the merge path.
Key points:
- Merge queues: SubmitQueue tests combinations of changes before landing to keep main healthy, but human-arrival-rate designs become bottlenecks under agent-generated PR floods.
- Cutting feedback cost: incremental builds, caching, sharding, and parallelism (illustrated with a Rust example comparing full rebuilds vs testing only affected work).
- Repair loops: agents need durable context from logs and failures so they don't repeatedly rediscover the same problem.
- Approval boundaries: policies should match change risk, separate authorship from approval, bind decisions to the exact tested commit, and record overrides with reasons and audit evidence.
Takeaway: measure the whole path from proposed change to accepted change—including recovery and accountability—before equating more generated code with more delivered value.
More from coding & agent
- Sentry CEO: AI agents still write messy code but are getting better at managing it — zeeg · 2026-10-15
- Omarchy bug: missing rtkit causes Chrome/Electron apps to get SIGKILLed — fix inside — mark_k · 2026-10-15
- 2,611 real interviews show rater bias predicts AI agent scores better than actual performance — Exp_Mark · 2026-10-15
- Google's DuoDiscover tops NVIDIA SOL-ExecBench, 2.68x faster than production TPU kernels — lateinteraction · 2026-10-15
- Agent stack check: harnesses, OTel telemetry and MCP are settled — evals remain a mess — _ScottCondron · 2026-10-15
- Claude Code's safety guardrails refuse even non-sensitive env var copies, user complains — Promptmethus · 2026-10-15