Data Engineering · Career Roadmap

Data Engineer 2027: The Real Stack And the Roadmap to It

Every job post lists twenty tools. This session cuts it to the stack that actually gets hired in 2027 — and the order you learn it in.

Most people learning data engineering are collecting tutorials, not building a career. They learn Spark before they can write a window function, or chase a new tool every month because a job post named it. This session replaces that with a sequence: six stages, and for each one what to learn, what to skip, and how you prove it. Foundations that never expire — Python, advanced SQL, Linux, Git. Moving and modelling data with Airflow orchestration and dbt transforms. Scale and streaming with Spark and Kafka, including the cases where batch is simply the right answer. The lakehouse era — Snowflake, Databricks, DuckDB, Trino. Shipping like an engineer — Docker, Kubernetes, Terraform, CI/CD and cost control. Then getting hired: the two portfolio projects that matter, the resume, and the interview loop. We also talk honestly about what AI changes in this role and what it does not. It is for analysts and SQL folks moving into data engineering, students planning the next six months, and anyone who has been learning for a year without a project they can defend. Bring your questions — the session ends with live Q&A.