Data Engineer (Python)

Das ist der Job

Role Rapidly validate new data initiatives end-to-end—without sacrificing adoptability.

Darum lohnt es sich

On Innovation, you’ll prototype representative connectors and pipelines (batch + streaming), generate credible performance/operability readouts, and ship a handoff package that Foundation or a delivery team can productize. What We Offer Modern data stack with real constraints: Kafka + NiFi + lakehouse + distributed SQL + Kubernetes.

High leverage: your prototypes become blueprints multiple teams reuse and productize. Data Engineer (Python) Company Orcrist builds the Orcrist Intelligence Platform (OIP), a Kubernetes-based data intelligence system delivered as SaaS or self-hosted/on-prem (including air-gapped deployments).

We run streaming and batch pipelines that power search, ML enrichment, and investigative workflows for mission-critical customers. What you'll do Prototype ingestion and connector patterns (batch + streaming) using NiFi, Kafka, Kafka Connect/Streams, and CDC approaches.

Design “prototype-grade but adoptable” schemas and data models with clear semantics and evolution discipline. Build incremental lakehouse datasets (Hudi/Iceberg/Delta patterns) and produce queryable outputs for realistic latency/throughput evaluation. Bake in data quality and provenance mindset early (checks, metadata hooks, operability basics).

Containerize and deploy prototypes on Kubernetes; deliver minimal runbooks/configs that make adoption straightforward. Produce adoption artifacts: schemas, reference implementations, technical design notes, and an integration backlog. About You 3+ years data engineering experience (level dependent) with real pipeline delivery beyond ad-hoc scripts.

Strong Python + SQL; comfortable building transformations, validation tooling, and pipeline glue code. Practical streaming/CDC fundamentals (ordering, duplication, replay, idempotency) and Kafka ecosystem experience. Familiar with lakehouse/storage and query layers (e.g., Hudi/Iceberg/Delta, Trino/Hive/Postgres) and how to make datasets usable.

Comfortable working in Kubernetes/container environments and documenting decisions clearly. Eligible to work in Germany; EU/NATO citizenship preferred and export-control screening applies. Nice‑to‑haves Great Expectations or similar data quality tooling; metadata/lineage platforms (OpenMetadata/DataHub/Atlas).

Experience shipping in on-prem or air-gapped environments; governance/policy awareness for regulated customers. German language (B1+) and/or experience with OSINT/GEOINT/multi-INT data shapes. Remote-first in Germany with regular Berlin prototyping sprints, 30 days vacation, equipment & learning budget.