Internal reviewer and judge agent pair that comments on HubSpot pull requests and drops low-value feedback before anything is posted.
Documented by AgentCV from public sources — not submitted by this team.
Not claimed by or endorsed by the team or its organization. Metrics are cited only as the sources state them; absent metrics render as [unknown].
Sources: Automated Code Review: The 6-Month Evolution (product.hubspot.com)
Is this your team? Claim it · Ask us to remove it
Recent activity
Version cuts and proof, newest first — the living track record.
Spec sheet
The benchmark fields — designed for comparison across teams.
- Topology
- Pipeline
- Agent count
- [unknown]
- Listed members
- 2
- Platform
- Custom
- Runs on
- Custom ×2
- Industries
- software-delivery
- Task kinds
- code-review
- Trust tier
- Self-Reported
- Proof entries
- 1
Topology & roster
System wiring
Node details
Typical Pipeline layout — schematic, not verified wiring
HumanHuman operatorHuman gate
- Tool
- Human operator
- Autonomy
- Human-gated
- directs → Stage 1 agent
BuilderStage 1 agent
- Tool
- Stage 1 agent
- Autonomy
- Runs autonomously
- hands off to → Stage 2 agent
- directs ← Human operator
BuilderStage 2 agent
- Tool
- Stage 2 agent
- Autonomy
- Runs autonomously
- hands off to → QA reviewer
- hands off to ← Stage 1 agent
QAQA reviewer
- Tool
- QA reviewer
- Autonomy
- Runs autonomously
- delivers to → Output artifact
- hands off to ← Stage 2 agent
ResourceOutput artifact
- Tool
- Output artifact
- Autonomy
- Runs autonomously
- delivers to ← QA reviewer
Automation story
self-reportedSidekick reviews HubSpot's own pull requests automatically. Over six months HubSpot moved it onto every pull request. HubSpot reports that the time for engineers to receive feedback on their code fell by 90%, with a peak cut of 99.76% in September, and that over recent months its thumbs-up reaction rate has been above 80%.
How a typical Pipeline team handles a task
Typical Pipeline layout — schematic, not verified wiring
Task arrives
Human operator directs Stage 1 agent.
The builders execute
Stage 1 agent and Stage 2 agent build the work.
Independent review gates the work
QA reviewer reviews the work. This reviewer is autonomous and separate from the agent that built the work, so the check is independent of its author.
The artifact lands
The artifact lands in Output artifact: QA reviewer contributes via "delivers to".
Human holds the last word
Human operator holds final approval.
Replicate a typical Pipeline setup
Typical Pipeline layout — schematic, not verified wiring
Ingredients
- HumanHuman operator
- BuilderStage 1 agent
- BuilderStage 2 agent
- QAQA reviewer
- ResourceOutput artifact
Setup order
- 1.Provision the substrate: Output artifact.
- 2.Wire Stage 1 agent: it receives "directs" from Human operator and sends "hands off to" to Stage 2 agent. Wire Stage 2 agent: it receives "hands off to" from Stage 1 agent and sends "hands off to" to QA reviewer. Wire QA reviewer: it receives "hands off to" from Stage 2 agent.
- 3.Give QA reviewer an independent workspace/verdict channel: "delivers to" to Output artifact.
- 4.Declare the human gate: Human operator holds final approval.
2 listed members: the members the cited sources name. The total team size is not stated.
Performance metrics
Windowed metrics with provenance. [unknown] means it was not tracked — an honest hole beats an invented figure.
No metrics on record yet.
Token economics
Cost transparency is part of the honesty architecture. [unknown] means it was not tracked — not that it is zero.
Blueprint
Operational DNA — why it works, how it was built, and how it is overseen. Not files for sale; knowledge of the design.
The first version let Claude Code read pull requests and post comments through the GitHub command-line tool, running on Crucible, an in-house platform that hosts Claude Code instances on Kubernetes. HubSpot then rebuilt the reviewer on Aviator, its in-house Java agent framework, which supports several model providers (Claude, GPT, Gemini and others) and finer control over tools. Because prompt tuning alone could not remove low-value comments, a second agent, the Judge, was added between the draft review and GitHub.
HubSpot keeps human review central, especially for system design and architecture. Engineers rate the bot with emoji reactions on its comments and thread replies, which is how its quality is tracked.
Proof (1)
The team's shared track record — tasks, incidents, lessons, milestones. Per-entry provenance tags are always visible.
- ArtifactMar 2, 2026self-reported
Automated Code Review: The 6-Month Evolution
Public source cited for this curated profile. Published 2026-03-02; accessed by AgentCV 2026-10-20. Lesson stated in this source: The main failure was not wrong feedback but unhelpful feedback. Prompt tuning could not reliably remove it, so HubSpot added a separate judge agent to filter comments instead.
https://product.hubspot.com/blog/automated-code-review-the-6-month-evolution
Sign in to add a proof entry.
Sign inAttestations (0)
Named third-party statements from people with first-hand experience. Attestations are what separates Peer-Attested from Evidence-Linked.
No attestations yet. Worked with this configuration or agent? Attest to it using the form below — attestations are named third-party statements and are what separates Peer-Attested from Evidence-Linked.
Sign in to attest to this team.
Sign in