Miss Irungu

Built a real-time Change Data Capture pipeline using AWS DMS streaming from RDS into Snowflake, reducing latency from 24 hours to under 5 minutes. Implemented Databricks lakehouse architecture, plus AWS Lambda and Kinesis event pipelines with schema validation and PII masking for governed ingestion.

Designed and delivered data engineering training for 50+ professionals covering Apache Airflow, Python orchestration, and cloud pipelines. Built AWS-based training environments (S3, RDS, IAM) to simulate production data systems and guide end-to-end pipeline development.

Designed cross-system data integration frameworks to support structured data exchange between partner organizations. Built Python-based automated reporting pipelines and defined data architecture standards to ensure consistency, traceability, and governance compliance.

Built Airflow-based orchestration pipelines for automated data movement and transformation workflows, standardizing reusable pipeline design patterns across teams. Simulated production-grade AWS environments and mentored engineers on ingestion and orchestration workflow design.

Managed data preparation pipelines for AI model training, ensuring dataset quality and consistency requirements. Coordinated engineering and labeling stakeholders, improved data cleaning to enhance training accuracy, and documented lineage, transformation logic, and validation processes for auditability.

Built large-scale data validation pipelines for machine learning datasets to meet production quality standards. Developed Python ETL frameworks and automated filtering to prevent corrupted data from entering training pipelines while optimizing cloud workflows with structured logging and performance-oriented batching/indexing.

Bachelor's degree in Computer Science from the University of the People.