Why Data Engineers Are Becoming AI Engineers

Alejandro Aboy, a Senior Data and AI Engineer at Workpath, makes a practical case that the two roles are merging – roughly 80-90% of his current work touches AI, but he can’t cleanly separate it from data architecture. His hierarchy is specific: data modeling accounts for 80% of impact, RAG is essentially the ETL of AI, and writing precise column descriptions doubles as context engineering for LLM systems. The advice to build MCP integrations for frequently-used tools and treat AI output as code to critique rather than accept maps directly to how teams using Claude Code and dbt are actually working now. The one thing AI still lacks, he argues, is instinct – knowing when not to act – which remains the real edge for experienced data engineers navigating production risk.