Dmytro K.

* Deployed and optimized ML pipelines on AWS (Lambda, EMR, EC2, EKS, S3) for structured and unstructured data at scale. * Designed CI/CD pipelines with GitHub Actions + Docker/Kubernetes, automating deployments of ML models into production. * Implemented ML monitoring solutions (Grafana, CloudWatch, Datadog), improving observability and uptime. * Collaborated with data scientists to package, containerize, and deploy computer vision and NLP models to AWS EKS.

* Delivered end-to-end Spark pipelines on Databricks and AWS, handling petabyte-scale data for personalization and analytics. * Designed batch + streaming ETL jobs with PySpark, Airflow, and Delta Lake, reducing data latency from 12h to near real-time. * Built data catalogs and metadata management solutions improving data discoverability and governance. * Integrated Kafka, SQS, and Kinesis for event-driven architecture and streaming ingestion.