Data Engineer

• Build and run pipelines that turn raw records into datasets used by Firmable's product and agents
• Implement SQL, dbt, Airflow, and Snowflake data pipelines
• Integrate LLM extraction, enrichment, and matching steps into pipelines processing millions of records per day
• Create structured outputs, retries, and human-review paths for model workflows
• Build evaluation sets and regression tests to measure model-step accuracy
• Route simple decisions to lower-cost models and complex tasks to stronger LLMs
• Track token costs per pipeline
• Deduplicate company and people records across covered markets
• Build self-recovering Airflow DAGs and budget-conscious Snowflake models
• Use coding agents such as Claude Code, focusing engineering time on design, review, and evaluations
• Log every LLM call with prompt version, model, and cost
• Ship improvements through weekly releases moving toward daily releases

• 3–5 years in data engineering
• Strong Python and SQL
• Production dbt and Airflow experience
• Experience with a cloud warehouse; Snowflake preferred
• Nice to have: Logfire or OpenTelemetry
• Nice to have: embeddings or fuzzy matching for entity resolution
• Nice to have: fine-tuning small models
• Nice to have: AWS and PostgreSQL
• Nice to have: Spark or streaming
• Nice to have: GDPR/SOC2
• Experience with LLM extraction, enrichment, and matching at production scale is implied by the role context
• Ability to use AI coding tools daily and show work shipped with them
• Production pipeline ownership, including fixing failures
• Experience working with millions or billions of records rather than only thousands of rows
• Experience building pipelines in code rather than only drag-and-drop ETL tools
• Ability to work independently without detailed specifications, fixed hours, or close supervision

• Competitive base and meaningful equity
• Fully remote work arrangement
• Equal opportunity employer