Analytics Engineering in the Modern Data Stack

Analytics Engineering in the Modern Data Stack — Lightweight, Open-Source, and End-to-End

The data world for the past decade has been obsessed with “big data”, but most real organizations don’t actually have it. What every organization needs is fast, reliable, and understandable analytics workflows powered by tools that are flexible, affordable, and friendly to both engineers and analysts.

In this talk, we’ll explore a modern, open-source data stack purpose-built for today’s analytics engineering workflows: Dagster for orchestration, dlt for ingestion and EL, dbt for transformation, DuckDB as an embedded warehouse, and Metabase for lightweight BI. Together, these tools create a seamless, local-first pipeline that’s powerful enough for production yet simple enough for rapid iteration and learning.

We’ll break down the evolving role of the analytics engineer, how it fits alongside analysts, data engineers, and data scientists, and why engineering best practices have become essential across all data roles. We’ll also discuss the industry shift from big data back to small data, and how embedded compute engines like DuckDB are reshaping how teams build pipelines and deliver insights.

The session a live walk-through of the entire stack in action, showing how these tools work together to create a clean, maintainable, modern analytics workflow.

Food and networking at 5:30, presentation at 6:00

Sponsored by Talestry