About the role 

We're seeking a seasoned engineer to design, operate, and scale our workflow orchestration platform with a primary focus on Apache Airflow. You'll own the Airflow control plane and developer experience end-to-end—architecture, automation, security, observability, and reliability—while also evaluating and operating complementary schedulers where appropriate. You'll build automation infrastructure and partner across data, trading, and engineering teams to deliver mission-critical pipelines at scale. 

 

What you’ll do 

• Architect, deploy, and operate production-grade Airflow on Kubernetes including all components and user application dependencies, with focus on upgrades, capacity planning, HA, security, and performance tuning 
• Operate a multi-scheduler ecosystem: determine when to use Airflow, distributed compute schedulers, or lightweight task runners based on workload requirements; provide unified developer experience across schedulers 
• Build automation infrastructure: Terraform modules and Helm charts with GitOps-driven CI/CD for environment provisioning, upgrades, and zero-downtime rollouts 
• Standardize the developer experience: DAG repo templates, shared operator libraries, connection and secrets management, dependency packaging, code ownership, linting, unit testing, and pre-commit hooks 
• Implement comprehensive observability: metrics collection, dashboards, distributed tracing, SLA/latency monitoring, intelligent alerting, and runbook automation 
• Enable resilient workflow patterns: build idempotency frameworks, retry/backoff strategies, deferrable operators and sensors, dynamic task mapping, and data-aware scheduling 
• Ensure reliability at enterprise scale: architect and tune resource allocation (pools, queues, concurrency limits) to support high-throughput workloads; optimize large-scale backfill strategies; develop comprehensive runbooks and lead incident response/postmortems 
• Partner with teams across the organization to provide enablement, documentation, and self-service tooling 
• Mentor engineers, contribute to platform roadmap and technical standards, and drive engineering best practices 

Required qualifications 

• 5–8+ years building/operating data or platform systems; 3+ years running Airflow in production at scale (hundreds–thousands of DAGs and high task throughput). 
• Deep Airflow expertise: DAG design and testing, idempotency, deferrable operators/sensors, dynamic task mapping, task groups, datasets, pools/queues, SLAs, retries/backfills, cross-DAG dependencies. 
• Strong Kubernetes experience running Airflow and supporting services: Helm, autoscaling, node/pod tuning, topology spread, network policies, PDBs, and blue/green or canary strategies. 
• Automation-first mindset: Terraform, Helm, GitOps (Argo CD/Flux), and CI/CD for platform lifecycle; policy-as-code (OPA/Gatekeeper/Conftest) for DAG, connection, and secrets changes. 
• Proficiency in Python for authoring operators/hooks/utilities; solid Bash; familiarity with Go or Java is a plus. 
• Observability and SRE practices: Prometheus/Grafana/StatsD, centralized logging, alert design, capacity/throughput modeling, performance tuning. 
• Data platform experience with at least one major cloud (AWS/Azure/GCP) and systems like Snowflake/BigQuery/Redshift, Databricks/Spark, EMR/Dataproc; strong grasp of IAM, VPC networking, and storage (S3/GCS/ADLS). 
• Security/compliance: SSO/OIDC, RBAC, secrets management (Vault/Secrets Manager), auditing, least-privilege connection management, and change control. 
• Proven incident leadership, runbook creation, and platform roadmap execution; excellent cross-functional communication. 

Nice to have 

• Experience operating alternative orchestrators (Prefect 2.x, Dagster, Argo Workflows, AWS Step Functions) and leading migrations to/from Airflow. 
• OpenLineage/Marquez adoption; Great Expectations or other data quality frameworks; data contracts. 
• dbt Core/Cloud orchestration patterns (state management, artifacts, slim CI). 
• Cost optimization and capacity planning for schedulers and workers; spot instance strategies. 
• Multi-region HA/DR for Airflow metadata DB; backup/restore and disaster drills. 
• Building internal developer platforms/portals (e.g., Backstage) for self-service pipelines. 
• Contributions to Apache Airflow or provider packages; familiarity with recent AIPs/Airflow 2.7+ features.