Senior Data Engineer (Snowflake)

Main requirements

Snowflake Experience

The person should be proactive, well-organized, and an effective communicator, able to independently drive tasks forward, keep processes under control, and communicate clearly with the team and stakeholders.

Requirements

Senior Data Enginer

Snowflake — core development

Writing and maintaining production views, secured views and stored logic in Snowflake

Understanding of role-based access control: users, roles, GRANTs, warehouse and database entitlement models

Enough query-tuning instinct to build views over large fact tables without wrecking performance — partition pruning, clustering awareness, avoiding SELECT *

Microsoft SQL Server

T-SQL development

Migrating legacy tables, views and stored procedures to Snowflake

Previous ETL Tooling Experience

Bonus Points

BI and semantic layer

Sigma Computing — connection setup, workbooks, embedding, AI/Cortex configuration. We are rolling Sigma out now and retiring Tableau Desktop.

Tableau — data source management, extract vs live strategy, extract scheduling and dependency mapping

Documentation and code review

Producing data-flow and architecture diagrams

Reviewing SQL/pipeline code to a standard — we have a code review backlog in other areas because there aren’t enough qualified SQL reviewers. A contractor who can act as a second reviewer is disproportionately valuable.

Azure DevOps (aka ADF Pipelines sync’d with Github)

CI/CD pipelines for database and ADF artefacts, service principal / client secret configuration, release troubleshooting

External API ingestion

Consuming and loading third-party APIs into the warehouse (e.g. FX/exchange rate feeds), including auth and error handling

Snowflake Git integration — connecting a Git repository to Snowflake, versioning database objects and deploying from a branch, plus general source-control discipline for SQL and pipeline code

Snowflake Streamlit — building lightweight in-warehouse data apps as an alternative to a full BI workbook where the audience is small or the use case is operational

Python for data engineering and Snowpark

Snowflake Cortex / LLM-in-warehouse features

dbt or equivalent transformation tooling (not in use today, but relevant if we modernise)

Working attributes we need

Comfortable being the second pair of hands on a one-person team — takes a ticket end-to-end without daily direction

Documents as they go; we are explicitly trying to reduce single-person key-man risk

Cost-aware: several of our decisions are spend-driven, not purely technical

English - Andvanced