Vision Data Platform Engineer - Scalable Image Pipelines

Qualifications Bitte lesen Sie die Informationen in dieser Stellenausschreibung sorgfältig durch, um genau zu verstehen, was von potenziellen Bewerbern erwartet wird. Experience designing and managing ETL/ELT processes for reliable ingestion, transformation, and delivery of structured and unstructured data. Strong production experience with Python and SQL Strong hands‑on experience with AWS, especially services relevant to data platforms such as S3, IAM, CloudWatch, SQS/SNS, ECS/EKS, Lambda, and RDS/PostgreSQL Experience designing and operating production‑grade data pipelines for both structured and unstructured data Experience with object storage and large binary assets such as images, documents, media, or similar unstructured data Experience with workflow orchestration tools such as Prefect, Dagster, or Airflow Experience with Postgre SQL or comparable systems for operational metadata, integrating REST APIs and building automation around external platforms or internal services Experience implementing data validation, data quality controls, observability, and monitoring in production systems Your Tasks Design and build scalable data pipelines that can reliably handle high volumes of clinical images and metadata coming from hospital and application workflows. Create strong data validation and quality processes to ensure images and metadata are complete, accurate, compliant, and easy to trace, while also handling duplicates, failures, and reprocessing when needed. Develop the underlying data architecture for large image assets and operational metadata, including storage design, canonical data models, lineage, and curated datasets that can support machine learning use cases. xhdzyng Build and manageend-to-end workflows that take data from ingestion through validation, pre‑annotation, annotation‑platform integration, curation, and final dataset release in a reliable and measurable way. Work closely with annotation and ML teams to deliver high-quality, versioned, and traceable datasets, while maintaining strong monitoring, auditability, and compliance‑aware data handling across the platform. #J-18808-Ljbffr