Whenever you're ready, walk me through how you'd approach this.
Describe the process of loading and processing a massive, multi-terabyte file.
Explain how you would choose storage and compute technologies, split the file for parallel processing, validate and transform records, and write the results safely. Cover memory constraints, retries, checkpointing, malformed records, observability, and how you would rerun or backfill the process without duplicating data.