Reviewed by Aditya Kumar · Last reviewed 2026-03-24
**Approach**: PySpark—spark.read.json() or from_json() for nested. Define StructType for type safety. Flatten, explode arrays. Example: df.select(explode('items')).groupBy('item.category').count(). from_json(col, schema) for nested. **Scale**: mode PERMISSIVE with...
Approach: PySpark—spark.read.json() or from_json() for nested. Define StructType for type safety. Flatten, explode arrays. Example: df.select(explode('items')).groupBy('item.category').count(). from_json(col, schema) for nested.
Scale: mode PERMISSIVE with columnOfCorruptRecord for malformed; partitioning for large. Pandas json_normalize for smaller.
Flipkart Context: High-volume event JSON; schema evolution; handle corrupt records gracefully.
Pro-Move: 'We use schema registry for JSON; corrupt records go to DLQ. Partition by date for 100GB/day ingest.'
According to DataEngPrep.tech, this is one of the most frequently asked General/Other interview questions, reported at 1 company. DataEngPrep.tech maintains an editor-reviewed database of 1,863 data engineering interview questions across 7 categories.