awesome-copilot/skills/phoenix-evals/references/experiments-synthetic-python.md

Experiments: Generating Synthetic Test Data

Creating diverse, targeted test data for evaluation.

Dimension-Based Approach

Define axes of variation, then generate combinations:

dimensions = {
    "issue_type": ["billing", "technical", "shipping"],
    "customer_mood": ["frustrated", "neutral", "happy"],
    "complexity": ["simple", "moderate", "complex"],
}

Two-Step Generation

1. Generate tuples (combinations of dimension values)
2. Convert to natural queries (separate LLM call per tuple)

# Step 1: Create tuples
tuples = [
    ("billing", "frustrated", "complex"),
    ("shipping", "neutral", "simple"),
]

# Step 2: Convert to natural query
def tuple_to_query(t):
    prompt = f"""Generate a realistic customer message:
    Issue: {t[0]}, Mood: {t[1]}, Complexity: {t[2]}
    
    Write naturally, include typos if appropriate. Don't be formulaic."""
    return llm(prompt)

Target Failure Modes

Dimensions should target known failures from error analysis:

# From error analysis findings
dimensions = {
    "timezone": ["EST", "PST", "UTC", "ambiguous"],  # Known failure
    "date_format": ["ISO", "US", "EU", "relative"],   # Known failure
}

Quality Control

• Validate: Check for placeholder text, minimum length
• Deduplicate: Remove near-duplicate queries using embeddings
• Balance: Ensure coverage across dimension values

When to Use

Use Synthetic	Use Real Data
Limited production data	Sufficient traces
Testing edge cases	Validating actual behavior
Pre-launch evals	Post-launch monitoring

Sample Sizes

Purpose	Size
Initial exploration	50-100
Comprehensive eval	100-500
Per-dimension	10-20 per combination