Analytics Engineer

We are seeking an Analytics Engineer to own the semantic layer and data mart strategy that connects our raw operational and financial data to the business teams who depend on it. A significant portion of your work will focus on modeling data replicated from SAP, our financial and other operational platforms into Snowflake building clean, tested, and well-documented dbt models that serve Hex dashboards and Snowflake semantic models queryable by LLMs.

• 3+ years in an Analytics Engineering, Data Engineering, or BI Engineering role delivering production dbt models at scale

You are comfortable using AI coding assistants (Claude Code, ChatGPT Codex, GitHub Copilot) to accelerate data model development, and you apply critical judgment to every line of AI-generated output. You understand that speed and quality are not in conflict when LLM tooling is used responsibly.

\n

\n

\n

SAP-to-Snowflake Data Modeling

\n

Data source ownership: Design and maintain the staging and intermediate dbt model layers for financial, operational, and asset-management data replicated from SAP (ERP) into Snowflake

\n

Source-system partnership: Collaborate with SAP and integration teams to understand table structures, key relationships, and business rules translating them into well-documented, testable dbt models

\n

Reliability & integrity: Implement incremental load strategies, referential integrity tests, null checks, and accepted-value constraints for all SAP-sourced models

\n

\n

Revenue & Asset Performance Modeling

\n

Data model ownership: Own the dbt models supporting power plant revenue calculations and budget variance analysis

\n

Asset performance marts: Build operational data marts integrating real-time and historical generation data, availability metrics, and O&M cost data for renewable asset evaluation

\n

Business alignment: Work closely with trading and risk teams to validate that margin and performance models reflect how the business measures and reports financial outcomes

\n

\n

Semantic Layer & LLM-Ready Data Models

\n

Semantic model development: Build and maintain Snowflake semantic models (definitions that allow LLMs and BI tools to query business metrics without ad-hoc SQL

\n

Machine-readable design: Structure mart-layer models with consistent grain, clear naming conventions, and rich metadata so that AI-assisted querying returns accurate, interpretable results

\n

Metric governance: Define and document business metrics, dimensions, and hierarchies in a format that is both human- and machine-readable, in close partnership with Data Engineers

\n

\n

Hex Dashboard Data Foundation

\n

Mart development: Build and maintain the data mart layer that serves Hex workbooks and dashboards consumed by Finance and Accounting teams, Asset managers, Energy traders, and executive stakeholders

\n

Performance optimization: Optimize mart models for Hex query patterns wide denormalized tables, pre-aggregated datasets, and semantic model hooks targeting sub-second dashboard response times

\n

Self-service enablement: Partner with analysts on Hex workbook design, providing clean, documented datasets and advising on best practices for self-service analytics

\n

\n

Data Quality & Governance

\n

Testing frameworks: Implement comprehensive dbt tests (generic and data tests, singular tests, dbt-expectations) with clear severity thresholds

\n

Standards & lineage: Establish and enforce naming conventions, data lineage documentation, and metadata standards across all analytical data assets

\n

Documentation: Define and maintain the data dictionary and mart-level documentation so both technical and business users can self-serve on definitions and model logic

\n

\n

AI-Assisted Development

\n

LLM productivity: Use AI coding assistants (Claude Code, ChatGPT Codex, GitHub Copilot etc.) to accelerate model scaffolding, SQL generation, test creation, and documentation while applying rigorous review to all AI-generated output

\n

Responsible AI practices: Participate in establishing team guidelines for secure, privacy-aware use of LLM coding tools, including managing data privacy boundaries and prompt injection risks in production environments

\n

\n

Cross-Functional Collaboration & Enablement

\n

Stakeholder engagement: Conduct data model walkthroughs and requirement workshops with business stakeholders across Finance, Accounting, Asset Management, Operations and Maintenance, Trading, and Credit risk teams

\n

Engineering partnership: Partner with Data Engineers on pipeline design and source-layer contracts; with Data Scientists and Analysts on feature-engineering-ready datasets

\n

Training & onboarding: Create documentation, data dictionaries, and training materials that enable analysts to use data marts and Hex workbooks independently

\n

\n