AI Data Engineering & ETL Pipeline Prompt Pack

AI Data Engineering & ETL Pipeline Prompt Pack — $1932 copy-paste AI prompts to design, build, test and run data pipelines — your real schemas, your exact tool versions, loads that are safe to re-run.AI tools write pipelines fast. Left alone they guess column names, use INSERT in tasks that get retried, read "the latest" partition instead of the run's window, skip late rows with a bare watermark, write Airflow 2 code for Airflow 3, and add tests that can never fail. This pack makes ChatGPT, Claude, Cursor, Claude Code and Codex work from your schemas, exact versions and pasted docs, mark unverified details VERIFY, design idempotent loads, keep personal data out of samples and logs, and hand you specs, SQL / YAML / Python, test plans, backfill plans and checklists you can check.WHAT YOU GET• PDF guide (59 pages, A4)• 32 Markdown prompts in 8 groups (foundations, ingestion, transformation, orchestration, data quality, schema & streaming, performance & cost, operations & governance)• 3 checklists (pipeline design, data quality & contracts, backfill & release)COVERS• dbt, Airflow, Dagster, Prefect, Spark, Polars, DuckDB, pandas, Kafka, Debezium, Iceberg, Delta Lake• Context block, request → spec, batch vs streaming, layers & naming• Source profiling, API and file ingestion, incremental extraction & watermarks, CDC• dbt models & YAML, incremental / microbatch models, facts & dimensions, SCD & snapshots, data-frame transforms• DAG authoring, idempotency & retries, backfills, data-aware scheduling & SLAs• Data tests, unit tests, data contracts, reconciliation, freshness & volume monitors• Schema evolution, partitioning & file sizes, streaming delivery semantics• Spark diagnosis, cost review, local dev loop, failure triage, PII & lineage, release readinessNOT INCLUDED• Application SQL & EXPLAIN tuning · CI YAML · incidents · RAG / LLM evals · vulnerability review• Fake benchmarks, cost savings or testimonialsTool behaviour checked against official docs on 30 October 2026. Independent; not affiliated with dbt Labs, the Apache Software Foundation, Dagster Labs, Prefect or Databricks.Author: Pluto · X @Mars_PeakulEdition 1.0 · October 2026