Founding Data Analytics Engineer

Broccoli AI is building a unified analytics foundation for trustworthy reporting and customer-facing insights. As the founding Data Analytics Engineer, you will own the data architecture, make key tooling choices, and ensure that every dashboard and analytic response is based on reliable, well-modeled data.

This role is focused on building an end-to-end data layer, from ingestion into the warehouse through modeling, documentation, and quality safeguards. You will also support ad-hoc deep dives when the team needs answers, partnering closely with engineering to keep data usable and stable.

What you will do

• Build and run ingestion pipelines: Set up reliable ingestion from all sources into ClickHouse, selecting the tooling and owning the end-to-end flow.
• Model clean, documented data: Transform raw feeds into curated tables, including entity resolution so the same customer is consistently represented across billing, support, and call-related data.
• Create the source-of-truth layer: Define canonical views and metric definitions that every dashboard and analysis uses.
• Make data human and AI-ready: Structure models, definitions, and documentation so both people and AI agents can query them accurately, and build internal tooling that helps others ask data questions with confidence.
• Maintain trust with quality controls: Implement freshness checks, quality tests, and alerts so pipeline issues are detected before customers notice.
• Run deep dives as needed: Support ad-hoc analysis, segment investigations, and partner questions.
• Work closely with engineering: Understand how systems store and produce data, including schemas, events, and architecture, and provide early input to ensure warehouse data is stable and easy to model.

What you bring

• 4–8+ years in data or analytics engineering, with experience building and operating production pipelines end to end, including being accountable when pipelines break.
• Strong SQL and solid Python, plus hands-on ETL tooling and orchestration (examples include Airbyte, Fivetran, Dagster, dbt or hand-rolled approaches).
• Experience with a columnar/OLAP warehouse; ClickHouse is preferred, with transferable experience from BigQuery, Snowflake, or Redshift.
• Data modeling depth: You have designed tables others query and care about what the numbers mean, not only that pipelines run.

Technologies

• ClickHouse, BigQuery, Snowflake, Redshift
• SQL, Python
• Airbyte, Fivetran, Dagster, dbt

Nice to have

• Self-directed experience, including being the first or only data person or building a data platform from scratch.
• ClickHouse-specific exposure, such as materialized views and performance tuning on event-scale data.
• Multi-source identity / entity resolution experience.
• Customer-facing or multi-tenant analytics exposure, including strict customer-level data isolation.
• B2B SaaS operational data experience (calls, bookings, jobs, billing) or CRM/field-service data such as ServiceTitan.

Location: San Francisco, CA (onsite)
Experience: 4 years minimum