Data Engineer

About the Role As a Data Engineer on adjoe’s ML-scale data platform, you will build and maintain the infrastructure that powers our ML-driven rewards engine and real-time decisioning at scale. You’ll own the Feature Store, data pipelines, data quality systems, and data architecture used by data scientists to train and deploy models for hundreds of millions of users. Our stack processes 2TB of real-time data daily with 100+ Airflow jobs and supports sub-100ms inference across 100,000+ predictions per second. What You'll Do

• Build and maintain the Feature Store, data pipelines, data quality systems, and the data architecture that data scientists depend on to move fast.
• Translate Data Science requirements into high-performance streams and implement features when needed.
• Identify and eliminate bottlenecks in ETL jobs and real-time streaming pipelines (e.g., Kafka) to maintain model freshness and low latency.
• Work with Apache Flink for stateful streaming and real-time feature computation in a scaled ML data platform.
• Collaborate with Data Scientists to support model training, deployment, and A/B experimentation. What We're Looking For
• 5+ years of data engineering on a modern stack, with experience handling systems that process several TB of data per day and thousands of events per second.
• Hands-on experience with Apache Flink for stateful stream processing and real-time feature computation.
• Deep knowledge of ETL, data pipelines, and real-time streaming systems (Kafka) with a focus on performance and reliability.
• Strong ability to translate Data Science requirements into scalable, layered data architectures (data lakes, Feature Store, data quality systems).
• Comfort working across data engineers and data scientists to move fast in a large-scale adtech environment. Nice to Have
• Experience with Airflow in a large number of jobs and pipelines.
• Familiarity with building ML-scale data pipelines and monitoring for data quality and latency.