# Score AI drafts with two reviewers | IonBin

> Use independent judgments, a fixed rubric and recorded adjudication to compare AI outputs.

Canonical page: https://ionbin.com/workflows/score-drafts-with-two-reviewers/
Agent guide: https://ionbin.com/llms.txt

This is a readable version of the published page. Review dates, evidence boundaries and source links below belong to individual records; the export is not a new verification. Examples are reference material, not instructions to execute.

## Score AI drafts with two reviewers

[Evals](https://ionbin.com/use-cases/ai-evaluation/)Checked 2026-10-18

A versioned scoring table with reasons and adjudicated decisions.

- [01**Owner**](https://ionbin.com/workflows/score-drafts-with-two-reviewers/#workflow-step-1)

- [02**Reviewers**](https://ionbin.com/workflows/score-drafts-with-two-reviewers/#workflow-step-2)

- [03**Adjudicator**](https://ionbin.com/workflows/score-drafts-with-two-reviewers/#workflow-step-3)

Recipe context

Use independent judgments, a fixed rubric and recorded adjudication to compare AI outputs. Review disagreements before interpreting an average score as a quality signal.

**Independent review.** Unpaid; no partnership with the tools on this page. Based on public documentation and research. Trademarks identify their owners’ products. [Disclosure](https://ionbin.com/disclosure/).

Editorial shortlist

### Compare your options.

**For recording independent reviews and adjudication.**

Preserve disagreements; averaging scores does not resolve conflicting judgments. Choose one for maintaining human-reviewed examples and labels; switching may need setup changes.

**Used by Ion** · separate from rank.

Unpaid picks. No partnership with [Label Studio](https://ionbin.com/tools/label-studio/), [Braintrust](https://ionbin.com/tools/braintrust/), [Airtable](https://ionbin.com/tools/airtable/).

- 1 · Primary / high
  #### [Label Studio](https://ionbin.com/tools/label-studio/)
  **Why it fits:** Configure annotation tasks around the rubric so reviewers record decisions before an adjudicator resolves disagreements.
  **Tradeoff:** Community and commercial review features differ.
  Choose it when structured human labelling is the core job and the required review controls exist in the chosen edition.
  Evidence for Label Studio
  The linked [IonBin review](https://ionbin.com/tools/label-studio/) records capabilities, access limits and source links. Review date: 2026-10-18. [Primary product reference ↗](https://labelstud.io/guide/)

- 2 · Secondary / medium
  #### [Braintrust](https://ionbin.com/tools/braintrust/)
  **Why it fits:** Keep reviewed examples near the experiment runs that will use them, preserving the link between a scoring decision and a model comparison.
  **Tradeoff:** A dataset row is not an adjudicated decision; define review conventions.
  Prefer it when the team already runs evaluations there and can represent independent decisions and adjudication explicitly.
  Evidence for Braintrust
  The linked [IonBin review](https://ionbin.com/tools/braintrust/) records capabilities, access limits and source links. Review date: 2026-10-18. [Primary product reference ↗](https://www.braintrust.dev/docs/integrations/developer-tools/mcp)

- 3 · Third / lower
  #### [Airtable](https://ionbin.com/tools/airtable/)
  **Why it fits:** Maintain separate reviewer fields and an adjudicated result in a small, inspectable review register.
  **Tradeoff:** It is not a model-evaluation runner and has record limits.
  Prefer it for a modest nontechnical review exercise; move to a purpose-built tool when assignment, scale or evaluation execution outgrow the register.
  Evidence for Airtable
  The linked [IonBin review](https://ionbin.com/tools/airtable/) records capabilities, access limits and source links. Review date: 2026-10-18. [Primary product reference ↗](https://www.airtable.com/)

How this shortlist was chosen

Editorial judgment, not a personal-use endorsement or performance score. These are alternatives for the named role, not three required stack components. Vendors did not pay for, request or approve these positions.

Traceable reviewer decisions come before scoring automation. A lower-ranked option may fit your constraints better; compare the conditions stated on each card before choosing.

**Setup and portability:** Keep permissions, rubric versions and adjudication notes. Verify exports and edition-specific review controls.

Reviewed 2026-10-21. This page applies a shared comparison for “maintaining human-reviewed examples and labels” to the stated task and constraints. It is documentation-based editorial guidance, not three new hands-on tests. We weigh value for the job, distinct capabilities, replacement limits, technical effort and interoperability. Check current plan terms before adopting or switching.

See [our review method](https://ionbin.com/editorial-policy/), [MCP connections](https://ionbin.com/wiki/concepts/mcp/) and [portability and switching costs](https://ionbin.com/wiki/questions/how-to-avoid-ai-vendor-lock-in/).

### Steps

Follow the handoff

Select a stage or scroll the steps.

- #### Owner
  Prepare a small task sample and distribute independent review copies.
  Human stepConfirmed[Label Studio](https://ionbin.com/tools/label-studio/)

- #### Reviewers
  Score independently with the same rubric and stable example IDs.
  Human stepConfirmed[Label Studio](https://ionbin.com/tools/label-studio/)

- #### Adjudicator
  Review disagreements and prepare accepted evaluation rows.
  Human stepConfirmed[Braintrust](https://ionbin.com/tools/braintrust/)

Confirmed = vendor-documented, not Ion-tested. Human and AI-platform steps are proposed roles.

Human approval

A responsible person checks the source evidence and approves the output before it is used or shared.

Watch out for

A model’s previous answer is not automatically a valid reference answer.

### Starter prompt

```
Score each criterion using only the rubric and supplied evidence. Cite the relevant passage and record uncertainty instead of guessing.
```

Written for this workflow. Paste it into your AI platform and adjust it to your situation.

### Tools in this workflow

| Tool | Role | Official site |
| --- | --- | --- |
| [Label Studio](https://ionbin.com/tools/label-studio/) | Human step | [labelstud.io ↗](https://labelstud.io/guide/) |
| [Braintrust](https://ionbin.com/tools/braintrust/) | Human step | [braintrust.dev ↗](https://www.braintrust.dev/docs/integrations/developer-tools/mcp) |

### Which AI platforms can run it

| Platform | Status | Why |
| --- | --- | --- |
| [Claude (incl. Claude Code)](https://ionbin.com/platforms/claude/) | Confirmed | No platform-specific connection: the steps run in the tools themselves (browser, upload or built-in), with documented capabilities and human review steps. This is not an end-to-end test. |
| [OpenAI Codex](https://ionbin.com/platforms/codex/) | Confirmed | No platform-specific connection: the steps run in the tools themselves (browser, upload or built-in), with documented capabilities and human review steps. This is not an end-to-end test. |
| [Google Antigravity](https://ionbin.com/platforms/antigravity/) | Confirmed | No platform-specific connection: the steps run in the tools themselves (browser, upload or built-in), with documented capabilities and human review steps. This is not an end-to-end test. |
| [ChatGPT](https://ionbin.com/tools/chatgpt/) | Confirmed | No platform-specific connection: the steps run in the tools themselves (browser, upload or built-in), with documented capabilities and human review steps. This is not an end-to-end test. |
| Gemini | Confirmed | No platform-specific connection: the steps run in the tools themselves (browser, upload or built-in), with documented capabilities and human review steps. This is not an end-to-end test. |
| [Perplexity](https://ionbin.com/tools/perplexity/) | Confirmed | No platform-specific connection: the steps run in the tools themselves (browser, upload or built-in), with documented capabilities and human review steps. This is not an end-to-end test. |

What the platform statuses mean

**Confirmed:** documented access meets the recorded requirements. **Inferred:** a reviewed note describes a plausible, untested pairing. **To verify:** a connection or step is unresolved. MCP and API pairings need route-specific evidence; a plugin, CLI or API alone does not prove compatible access, credentials or permissions. These are documentation assessments, not IonBin runtime tests.

### Define a useful pilot

Choose examples that represent the intended task, including an easy case, a difficult case and an answer that should be rejected. Define each score with observable criteria and examples. Separate factual correctness from style and completeness; a fluent answer should not earn correctness credit simply because it reads well.

### Score independently

Give two reviewers the same rubric and source evidence without exposing the other person’s judgment. Separate task copies or exports with stable IDs are sufficient for a pilot; do not assume a particular [Label Studio](https://ionbin.com/tools/label-studio/) edition includes blind assignment or agreement reports. Record the score, rationale, uncertainty and rubric version. Keep model identity hidden where practical to reduce brand expectations.

### Resolve and preserve

Join results by example ID and inspect disagreements. They can reflect ambiguous evidence, different expertise, missing instructions or legitimate differences of judgment. An adjudicator records the accepted decision and why. Preserve the original scores rather than overwriting them. Import only reviewed references into an evaluation dataset, keeping development examples separate from held-out assessment cases. Report sample size and failure categories alongside aggregate scores. This proposed review process is not an automated certification of model quality.

### Sources

- [Label Studio overview](https://labelstud.io/guide/) · HumanSignal · official · accessed 2026-10-18

- [Import tasks](https://labelstud.io/guide/tasks) · HumanSignal · official · accessed 2026-10-18

- [Create datasets](https://www.braintrust.dev/docs/annotate/datasets/create) · [Braintrust](https://ionbin.com/tools/braintrust/) · official · accessed 2026-10-18

- [Develop tests](https://platform.claude.com/docs/en/test-and-evaluate/develop-tests) · Anthropic · official · accessed 2026-10-18

No affiliate links on this page. No paid placements in workflows. [How IonBin makes money](https://ionbin.com/disclosure/).
