• or equivalent orchestration and transformation tools.
• Engineer and optimise the ClickHouse warehouse for sub-second query performance across all back-offices.
• Implement data contracts between back-office and the platform. Onboarding a new operator is a config change, not new tables, topics, or feature views.
• Build the feature-serving layer providing pre-computed features to AI agents at millisecond latency.
• Integrate with third-party databases, back-office APIs, and external systems (CRM, affiliates, acquisition platforms).
• Establish monitoring, alerting, and maintenance procedures including pipeline health checks, freshness monitoring, anomaly detection, and data contract SLA enforcement.
• Own CI/CD and infrastructure-as-code for data workloads.
• Collaborate with data scientists, agent engineers, BI developers, and infrastructure teams to translate data requirements into reliable, production-grade pipelines.

About You:

• 3+ years building and operating production data pipelines at scale, with hands-on experience across both streaming and batch paradigms.
• Expertise in Apache Kafka (or Amazon MSK): topic design, consumer group management, offset handling, schema registry operations, and production troubleshooting of lag, rebalancing, and throughput issues.
• Strong SQL and warehouse engineering skills: experience with columnar analytical databases (ClickHouse strongly preferred, or similar: Druid, BigQuery, Redshift).
• PySpark / Spark Streaming proficiency: writing transformation jobs that normalise, enrich, and enforce business rules on event streams. Experience with AWS Glue, Apache Airflow, or Apache NiFi is a strong plus.
• Data modelling discipline: ability to design normalised, multi-tenant schemas where tenant isolation is a filter, not a fork. Experience with data contracts and schema governance.
• CI/CD and infrastructure-as-code experience: automated testing of data pipelines, version-controlled deployments (CloudFormation, Terraform, or CDK), and familiarity with containerised workloads (ECS Fargate or Kubernetes).
• Data quality and observability mindset: experience implementing pipeline health monitoring, automated data validation (Great Expectations or equivalent), freshness checks, and anomaly detection.
•