Senior Software Engineer, Developer Platform

Company Introduction

Key Responsibilities

• Architect, deploy, and operate workflow orchestration platforms (e.g., Argo Workflows, Airflow) supporting simulation, ML training, data pipelines, CI/CD, and general workloads.
• Build internal platforms, abstractions, SDKs, and self-service tooling on top of orchestration engines.
• Operate workflow platforms at scale on Kubernetes across cloud (AWS) and on-prem data centers.
• Ensure reliability, performance, and cost efficiency through observability, queuing, retries, and resource optimization.
• Partner with ML, simulation, data, and infrastructure teams to understand workload requirements.
• Integrate workflow platforms with storage, data streaming, artifact registries, and CI/CD tooling.
• Establish best practices, templates, and documentation; mentor engineers.
• Handle user-impacting issues promptly with clear communication and durable solutions.

Required Qualifications

• Bachelor's degree in Computer Science, Engineering, or related field, or equivalent experience.
• 5+ years of hands-on experience in platform engineering, infrastructure, DevOps, or SRE roles.
• Significant experience with workflow orchestration platforms such as Argo Workflows, Airflow, or comparable systems.
• Strong software development skills in one or more languages: Python, Go, Java, or JavaScript/TypeScript.
• Solid understanding of Kubernetes and distributed systems.

Preferred Qualifications

• Expert-level experience operating Argo Workflows, Airflow, and/or other engines (e.g., Prefect, Dagster, Temporal, Kubeflow Pipelines, Flyte).
• Experience orchestrating ML training, simulation, or large-scale data and batch workloads, including GPU scheduling.
• In-depth Kubernetes experience (EKS, GKE, AKS, RKE2/Rancher) and cross-cluster orchestration.
• IaC tools proficiency, including Terraform, Pulumi, OpenTofu, or Ansible.
• Experience with data streaming and event platforms, including NATS JetStream, Kafka, Pulsar, or RabbitMQ.
• Familiarity with observability stacks: Prometheus, Grafana, Loki, OpenTelemetry, or comparable.
• Demonstrated ability to optimize workload cost and performance without compromising reliability.