Analytics Engineer

CTEs, window functions, multi-dialect — the primary AE skill

pandas, scripting, and warehouse APIs

Columnar storage, partitioning, compute costs

Metric definitions, KPI trees, business vocabulary

dbt projects live in Git — this is non-negotiable

dbt Copilot, Cortex AI, text-to-SQL — know the limits

Project structure, compilation, run/test/docs cycle

VIEW vs TABLE vs INCREMENTAL vs EPHEMERAL

not_null, unique, relationships, custom — CI-gated

Self-documenting models — publish to GitHub Pages

Reference data + SCD Type 2 tracking

Source freshness checks + clean staging layer

DRY transformations — Jinja in dbt context

dbt-utils, dbt-expectations — standard installs

The dominant methodology for analytical data modeling

Enterprise auditability — hubs, links, satellites

The core warehouse design patterns — trade-offs matter

SCD Type 1/2/3/6 — dbt snapshots implement Type 2

Operational vs analytical — the fundamental trade-off

Reading source schemas — prerequisite for staging

Pre-joined wide tables — when to use vs star schema

Most-used enterprise warehouse — deep knowledge required

GCP standard — ARRAY/STRUCT, partitioning, clustering

Delta Live Tables, Unity Catalog, dbt on Databricks

AWS ecosystem warehouse — serverless + RA3 nodes

Local dev standard — same SQL, no warehouse costs

The open table format standard — all major warehouses support it

Slow models cost money — EXPLAIN plans, clustering, caching

Trade-off fluency — architecture conversations with leadership

Define metrics once — computed consistently everywhere

Semantic modeling in Looker — dimensions, measures, explores

Open-source headless BI semantic layer

Separate metrics from visualisation — the 2026 architecture

Ownership, versioning, change management for metrics

One definition of revenue everywhere — the AE north star

Power BI, Tableau, Looker — connecting to the semantic layer

Python data quality framework — 50M+ downloads

Great Expectations patterns as native dbt tests

Open-source monitoring built on dbt — Slack alerts + UI

Enterprise observability — ML-driven anomaly detection

YAML-based quality checks — SodaCL for non-engineers

Freshness, completeness, accuracy commitments — encoded as checks

Source column changes without breaking downstream models

Formal API between data producers and consumers

Z-score, IQR, and ML-based data quality monitoring

End-to-end dbt project — the gold standard portfolio signal

Published docs site — the hiring manager's first click

dbt Certified Developer + SnowPro + GCP Data Engineer

Specific warehouse experience + dbt project link

SQL modeling questions + dbt architecture discussions