1.3.1 · Lesson 8 of 28
Data validation workflows
Validate that input data meets quality requirements. Check:
WHAT
Validate that input data meets quality requirements.
Check:
- completeness
- validity
- schema
- missing values
- duplicates
- unexpected values
- encoding
- data types
WHY
Poor data can produce poor FM responses.
AWS services
- AWS Glue Data Quality
- SageMaker Data Wrangler
- AWS Lambda
- Amazon CloudWatch
Example
S3 Dataset ↓ Glue Data Quality ↓ Pass? ── Yes → Processing │ No ↓ Quarantine / Alert
EXAM TRAP
Do not send invalid data directly into the FM pipeline if a quality gate is required.
---
Key points
- Poor data can produce poor FM responses.
Exam focus
Focus on requirement-driven choices and common architecture traps.