Golden Dataset
Comprehensive patterns for building, managing, and validating golden datasets for AI/ML evaluation. Each category has individual rule files in rules/ loaded on-demand.
Quick Reference
| Category | Rules | Impact | When to Use |
|---|
| Curation | 2 | HIGH | Content collection, annotation pipelines |
| Management | 2 | HIGH | Versioning, backup/restore |
| Validation | 1 | CRITICAL | Regression testing |
| Add Workflow | 1 | HIGH | 9-phase curation, quality scoring, bias detection, silver-to-gold |
Total: 6 rules across 4 categories. House thresholds and scars: references/ork-delta.md.
Curation
Content collection, multi-agent annotation, and diversity analysis for golden datasets.
| Rule | File | Key Pattern |
|---|
| Collection | rules/curation-collection.md | Content type classification, quality thresholds, duplicate prevention |
| Annotation | rules/curation-annotation.md | Multi-agent pipeline, consensus aggregation, Langfuse tracing |
Difficulty ladder, coverage floors, and duplicate thresholds: references/ork-delta.md.
Management
Versioning, storage, and CI/CD automation for golden datasets.
| Rule | File | Key Pattern |
|---|
| Versioning | rules/management-versioning.md | JSON backup format, embedding regeneration, disaster recovery |
| Storage | rules/management-storage.md | Backup strategies, URL contract, data integrity checks |
CI automation for backups is upstream's job; see "Upstream coverage" below.
Validation
Quality scoring, drift detection, and regression testing for golden datasets.
| Rule | File | Key Pattern |
|---|
| Regression | rules/validation-regression.md | Difficulty distribution, pre-commit hooks, full dataset validation |
Schema validation and duplicate detection are upstream's job (see "Upstream coverage"
below); the house thresholds they must enforce live in references/ork-delta.md.
Add Workflow
Structured workflow for adding new documents to the golden dataset.
| Rule | File | Key Pattern |
|---|
| Add Document | rules/curation-add-workflow.md | 9-phase curation, parallel quality analysis, bias detection |
Quick Start Example
async def validate_before_add(document: dict, source_url_map: dict) -> dict:
"""Pre-addition validation for golden dataset entries."""
errors = []
if "placeholder" in document.get("source_url", ""):
errors.append("URL must be canonical, not a placeholder")
if len(document.get("title", "")) < 10:
errors.append("Title too short (min 10 chars)")
if len(document.get("tags", [])) < 2:
errors.append("At least 2 domain tags required")
return {"valid": len(errors) == 0, "errors": errors}
Key Decisions
| Decision | Recommendation |
|---|
| Backup format | JSON (version controlled, portable) |
| Embedding storage | Exclude from backup (regenerate on restore) |
| Quality threshold | >= 0.70 quality score for inclusion |
| Confidence threshold | >= 0.65 for auto-include |
| Duplicate threshold | >= 0.90 similarity blocks, >= 0.85 warns |
| Min tags per entry | 2 domain tags |
| Min test queries | 3 per document |
| Difficulty balance | Trivial 3, Easy 3, Medium 5, Hard 3 minimum |
| CI frequency | Weekly automated backup (Sunday 2am UTC) |
Common Mistakes
- Using placeholder URLs instead of canonical source URLs
- Skipping embedding regeneration after restore
- Not validating referential integrity between documents and queries
- Over-indexing on articles (neglecting tutorials, research papers)
- Missing difficulty distribution balance in test queries
- Not running verification after backup/restore operations
- Testing restore procedures in production instead of staging
- Committing SQL dumps instead of JSON (not version-control friendly)
Running a dataset as an experiment
Curating a dataset is half the job; the other half is running something against it and scoring the
result. Both Langfuse SDKs ship a runner, and their shapes differ.
Python (SDK 4.x): see monitoring-observability/references/experiments-api.md.
JS/TS (SDK 5.x): @langfuse/client exposes the runner directly on a fetched dataset.
import { LangfuseClient } from "@langfuse/client";
const langfuse = new LangfuseClient();
const dataset = await langfuse.dataset.get("my-evaluation-dataset");
const result = await dataset.runExperiment({
name: "Retrieval quality",
task: myTask,
evaluators: [myEvaluator],
});
| Type | Scores | Use for |
|---|
Evaluator | one item | Per-example quality (faithfulness, relevance) |
RunEvaluator | the whole run | Aggregate assertions — pass rate, mean score, regression checks |
Evaluation | — | { name, value, comment?, metadata?, dataType?, configId? } |
A per-item Evaluator cannot see the other items, so anything comparative belongs in a
RunEvaluator. createEvaluatorFromAutoevals wraps an autoevals scorer instead of hand-writing
one, and RegressionError is thrown when a run regresses against a configured baseline — catch it
to fail CI on a quality drop rather than only on an exception.
Full JS surface: monitoring-observability/references/langfuse-js-v5.md.
Evaluations
See test-cases.json for 9 test cases across all categories.
Upstream coverage (do not restate)
House thresholds these must enforce: references/ork-delta.md.
Related Skills
ork:rag-retrieval - Retrieval evaluation using golden datasetork:monitoring-observability - Langfuse tracing patterns for curation workflowsork:testing-llm - Evaluation harnesses that consume golden datasetsork:testing-unit - Unit testing patterns and strategies
Capability Details
curation
Keywords: golden dataset, curation, content collection, annotation, quality criteria
Solves:
- Classify document content types for golden dataset
- Run multi-agent quality analysis pipelines
- Generate test queries for new documents
management
Keywords: golden dataset, backup, restore, versioning, disaster recovery
Solves:
- Backup and restore golden datasets with JSON
- Regenerate embeddings after restore
- Automate backups with CI/CD
validation
Keywords: golden dataset, validation, schema, duplicate detection, quality metrics
Solves:
- Validate entries against document schema
- Detect duplicate or near-duplicate entries
- Analyze dataset coverage and distribution gaps
