At SambacLab, I’m a Data Engineer working on the Equinor TEP client project. I designed a historical data backfill pipeline with Scala, Apache Spark, and Delta Lake that remains compatible with a separate Rust-based ingestion layer.
I implemented Apache Arrow schema compatibility checks between Spark and Rust pipelines, and built a graph-based processing pipeline to reconcile equipment reference data with a PostgreSQL/GraphQL graph store. Checkpointing lets multi-hour jobs resume without reprocessing completed entities.
At SambacLab, I also designed a distributed data platform for ingestion, storage, querying, metadata, and data lifecycle operations. Its processing components handle 1GB+ datasets, and its concurrent processing supports 100+ sessions while reducing real-time event latency by 40%.
I built and published pymycorr on PyPI, a Python client that retrieves table data through Apache Arrow IPC and exposes it to Pandas, Polars, and Jupyter. I’ve also provisioned GCP infrastructure and deployed Kubernetes workloads using Terraform, Helm, and ArgoCD.
