Superhuman AI: Decoding the Future The Future of Data Engineering... It's AI
37 snips
Nov 6, 2025 Explore how AI is revolutionizing data engineering, making it smarter and faster. Discover the potential of AI-generated pipelines and the shift from traditional ETL to adaptive systems. Learn about the critical importance of local and enterprise context for effective AI work. Hear insights on common pitfalls in AI data approaches and real-world challenges with data instrumentation. The hosts also discuss their innovative tools, such as Moostack, that empower developers to create production-ready data solutions with ease.
AI Snips
Chapters
Transcript
Episode notes
SQL Chat Is Easy — Data Plumbing Is Hard
- Natural language queries are table stakes; the hard work is getting high-quality data into a warehouse accessible to LLMs.
- Enterprise integration and ingestion are where experienced data engineers add the most value.
The Red Shoes Problem At Nike
- At Nike, answering "how many red shoes did we sell yesterday" required deep contextual rules like Pantone codes and timezone definitions.
- That example shows why embedding metadata and enterprise policies into LLM context is crucial.
Feed Local And Remote Context To Copilots
- Provide both local developer context (logs, code, dev server) and remote enterprise context (PII/GDPR policies) to your copilot.
- Embed governance and policy into the LLM context so it suggests safe transformations automatically.