About the role
Responsibilities:
- Design and operate batch and streaming data pipelines (ETL/ELT) at production scale.
- Build and maintain training-ready datasets for ML/AI feature pipelines, labeling workflows, and data quality for model training.
- Own the data warehouse/lakehouse stack (e. g., Snowflake, BigQuery, Databricks, Spark) and transformations (dbt).
- Orchestrate workflows (Airflow or equivalent) with strong monitoring and lineage.
- Implement data security and compliance: PII classification, encryption, access controls, retention, and governance (GDPR and similar).
- Partner with AI/data science teams to make data discoverable, trustworthy, and fast.
Requirements:
- 5-7 years in data engineering, ideally within a SaaS environment.
- Hands-on experience preparing data for model/ML training (not just BI/reporting pipelines).
- Strong SQL and Python; deep experience with a modern warehouse/lakehouse and Spark.
- Comfortable integrating data services with a Node.js / TypeScript platform (APIs, events, schemas).
- Pipeline orchestration (Airflow) and transformation (dbt) at scale.
- Demonstrated work under data security and compliance requirements.
- Strong data modeling and data quality fundamentals.
Bonus:
- Streaming (Kafka, Flink, Kinesis) and real-time feature stores.
- MLOps / feature store tooling (Feast, Tecton).
- Experience in a multi-tenant or regulated data environment.
- Non-negotiables: 5-7 years of SaaS data engineering hands-on ML/training data prep, data security, and compliance experience.