Responsibilities Own and evolve streaming data pipelines that power live inference and real-time model serving across Kraken's AI infrastructure Design and build feature stores that serve low-latency, high-reliability features to production ML models Implement and maintain streaming systems using RisingWave, Apache Flink, or Kafka Streams, selecting the right tool for the workload Partner with ML engineers and AI infra teams to define data contracts, feature schemas, and pipeline SLAs Drive pipelines toward real-time where batch exists today reducing latency from hours to seconds Ensure data quality, observability, and auditability across all streaming and feature engineering systems Contribute to inference pipeline tooling where data engineering and model serving intersect Evaluate emerging streaming and feature store technologies and shape the team's technical roadmap Requirements 5+ years in data engineering with at least 2 years focused on streaming systems in production Hands‑on experience with RisingWave, Apache Flink, Kafka Streams, or comparable stream processing frameworks Strong understanding of feature store design — online/offline consistency, point‑in‑time correctness, low‑latency serving Experience building data pipelines that feed production ML models or inference systems Proficiency in Python and/or Scala; SQL fluency required Familiarity with data quality frameworks, pipeline observability, and SLA ownership Comfortable operating in a fast‑moving, ambiguous environment embedded within an AI‑focused team. #J-18808-Ljbffr
