Erwan D.

Curious and tenacious: I love digging deep into code, tracing call chains, scouring documentation and issues, and hunting down the micro-optimization that changes everything. I am tasked with simplifying, stabilizing, and accelerating existing data platforms — and leaving the team with understandable and sustainable tools.

Experience

Active since 2019, I have worked across real-time back-end, heavy data pipelines, and high-performance rendering/processing. Typical successes: fixing collectors and batch jobs, resolving bottlenecks (partitioning, compaction, skew, memory), making streaming reliable, and industrializing on Kubernetes (clean images, CI/CD, secrets, storage). I navigate well between diagnosis and production deployment, with a real focus on readability, testing, and observability.

Key Skills

• Data engineering: Spark (batch / Structured Streaming), Kafka, Hive/Parquet, Iceberg/Delta, S3/MinIO

• Languages: Scala, Java, (Python for support)

• Platform: Kubernetes / Spark Operator, Docker, CI/CD (GitLab/GitHub), Grafana/Prometheus/ELK, Argo Workflow, Helm

Open-source & personal projects

I contribute and publish regularly: developing an InfluxDB connector for Trino, contributions to Argo Workflows. To expand my technical horizons, I also develop Flutter apps (an AI-powered news reader).

On the personal R&D side: a ray tracer in Scala, compression algorithms, an interpreter/VM in Scala Native, OpenCL bindings in Java, a C++/OpenGL 3D engine (GitHub: k3rnL). These projects feed my approach to performance, algorithms, and low-level architecture.