- Engage stakeholders directly to gather, clarify, and document project requirements.
- Translate requirements into architected data solutions: choose the right storage, pipeline, modeling, and delivery approach for each problem.
- Own testing end-to-end — unit tests, data quality checks, reconciliation, and integration tests before anything reaches production.
- Deploy solutions to production and monitor post-deployment health, iterating rapidly based on real-world feedback.
- Run parallel AI coding sessions (Claude Code, Cursor, Codex) across different facets of a pipeline simultaneously — orchestrate, verify, and integrate the outputs.
- Build and maintain context files (CLAUDE.md equivalents) for data projects that encode schema conventions, pipeline patterns, and institutional knowledge — making every future AI session smarter.
- Design verification loops: automated data quality checks, dbt tests, CI hooks, and pipeline monitors that give AI agents concrete feedback on correctness.
- Build MCP (Model Context Protocol) or equivalent integrations to connect AI agents directly to Snowflake, Amazon Athena, Postgresql, MySql, Power BI APIs, Salesforce, and internal tooling.
- Design and build complex, reliable data pipelines ingesting from AWS, Azure, Salesforce, MuleSoft, and multiple third-party APIs into our AWS Data Lake and Snowflake warehouse.
- Implement and evolve data models using Kimball methodology to support financial, operational, and clinical analytics.
GraphQLNode.jsPython+9 more