Data Platform Engineer

Job description

## About Hazel Hazel is the AI coworker for consumer brands. We connect to a brand's live data — Shopify, Klaviyo, Amazon, Meta, and dozens of others — and turn it into a teammate operators talk to every day. Ask "why did repeat purchase rate drop last week?" and 10 seconds later you get an answer, cited from live data. Hazel also takes action herself, proactively flagging issues and making changes directly in Shopify and the other systems she connects to, on the way to running full playbooks autonomously. We’ve raised more than $2M, are rapidly scaling and serving some of the biggest consumer brands including Bogg Bag, Ultra, and OneSkin. ## Why this role exists Hazel’s intelligence is built on top of a data warehouse meticulously tailored for consumer brands. We’re moving a massive amount of data and the owner of this layer plays a direct part in how users interact with Hazel Our data layer is tens of terabytes of data and thousands of data tables, and customers expect perfect sync reliability, new integrations to be built in days, and for every question Hazel asks to be correctly grounded in data. ## The stack Experience with our stack is a bonus, but similar experience working on problems at our scale is required: - dbt - Temporal - dltHub - Duckdb / Motherduck - Dagster ## What you'll do You own the data platform end to end — ingestion, transformation, orchestration, and the reliability of all three. As importantly, you’ll need to further develop our existing AI agents that write, build, and test new integrations end-to-end. Automating your job is the only way you will scale it. **Beyond that you’ll:** - Own data platform reliability as we scale - Optimize ingestion pipelines for low-latency data availability, ensuring consistent performance during peak seasonal surges. - Build our data transformation layer so Hazel has clean, predictable data to work with. - Automate your own job, especially the data integrations and transformations - Ship new source integrations end to end ## Who we're looking for - 4+ years building production data infrastructure, with real ownership of a warehouse someone else depended on. - Deep understanding of dbt and SQL. You have opinions about grain, incrementality, and data modeling. - Strong Python. Our ingestion layer is code you'll be writing, not a UI you'll be clicking. - You've owned an orchestrator in production — Dagster, Airflow, Temporal, Prefect, whatever es. - AI-native in practice. You use coding agents to do the work of a much larger team, and you can tell us where they helped and where they made things worse. - You write documentation well, because here it's a feature. - Comfortable being the only person who does this job, and equally comfortable making sure that stops being true. Nice to have: Experience with open table formats, e-commerce/DTC data (Shopify, Klaviyo, Amazon SP-API, ad platforms), and designing multi-tenant warehouses. ## Comp & benefits - $155–200K base - 0.3–0.5% equity - Top-tier health, dental, vision - 401(k) - Unlimited PTO - Hybrid in NYC ## How we hire - 15-min intro - Founder call - Systems design and case study - Onsite